How A College Education Divides American Voters
01:00:10 Share

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com. Nate, what are you doing? I'm making a sports bet.

Are you f***ing kidding me? What are you betting on? What are you betting on? I bet on the Memphis Grizzlies against the Brooklyn Nets. Which, what bet did you make? I'm not saying. Hello and welcome to the FiveThirtyEight Politics Podcast. I'm Galen Dweer.

There are just two weeks until Election Day, and according to our forecast, the race for the Senate is a dead heat. Democrats have a 55% chance of winning the chamber, to Republicans' 45% chance. Republicans have an 81% chance of winning the House.

we've also seen some movement in key governor's races. The race in Wisconsin has moved from lean Democratic to a toss-up, and the races in Michigan and Pennsylvania have moved from solid Democratic to likely Democratic. Today, we're going to ask whether it's a good or bad use of data to suggest that some of the recent shift in voters' preferences has to do with increased ad spending amongst Republican candidates. That may assume that political ads work in the first place, but do they?

We're also going to dig into a new poll of Latino adults, suggesting a quarter of Latinos are still undecided and another 17% don't plan to vote. When it comes to preferences between Democrats and Republicans, Democrats have half the advantage amongst Latinos they usually have in this poll. But, and how could we not ask it, is it a good or bad use of polling? And we often talk about the educational divide in American politics, but what do we really mean? Sports.

especially as we head into another election where we expect to be there to be a significant educational divide. What are the beliefs and preferences that come with a diploma or no diploma? And why is that the difference maker? We're going to start with the question of ad spending. And with me to discuss is editor in chief Nate Silver. Hey, Nate. Hey, everybody. Also here with us is senior elections analyst Nathaniel Rakich. Hey, Nathaniel.

Hey, Galen. And later on, we'll be joined by friend of the pod, Carlos Odio and FiveThirtyEighter, Monica Potts. Before we dive in, though, if things look or sound a little different today, it's because I am in Washington, D.C. Nate, you're going to be joining me later on today. And it's, of course, because we have our live show tomorrow, Tuesday, October 25th at 6th and I. Nate, are you excited?

I am. It's our first live show since COVID, right? I think it is. Yeah, absolutely. In fact, our last live show before COVID was in Washington, D.C. at Sixth and I. So we're going back to where we left off. I'm looking forward to it quite a bit, I'd say.

And that means we're going to be recording this week's model talk live in front of an audience. I'm excited. If you don't have your tickets yet, you still have the opportunity to get them. I'll put a link in the show notes. It's at seven o'clock tomorrow at six and I or if you're listening to this on Tuesday, it's today. So go get those tickets now. Nathaniel is going to pop by. Amelia is going to pop by. We got a lot going on. We're going to play some games. It's going to be really fun. All right. We have a lot to get into today. So let's get to it.

Ad Impact, an advertising intelligence firm that tracks ad spending, recently released numbers on the amount of money spent on airtime reservations for Senate campaign ads by the two parties for the final three weeks of the election. It showed Democrats spending about $29 million more on all Senate ads than Republicans.

They also broke down which races had the most money being sunk into them, with Pennsylvania, Georgia, and Nevada ahead by a large margin. So perhaps they've been looking at our forecasts. In response to these numbers, elections analyst Lasha Jane wrote on Twitter, quote, One major ray of hope for Democrats is that the ground gained by Republicans in statewide polling of late corresponded quite sharply with a heavy TV reservation advantage.

That is decidedly not the case in the last three weeks in the core battlegrounds. So Nathaniel, is this a good or bad use of data?

Yeah.

You know, people, especially in this age of partisanship, people, it's just hard to kind of persuade people. And so, you know, and in addition, you know, you kind of maybe Democrats have the the advertising advantage in the next two weeks right now. But also that can change. Right. You know, Republicans can buy more ads. Democrats could cancel some ads. I don't think we're at the point yet where the ads are like literally sold out. I haven't heard about that yet, at least so far.

So I'm much more inclined to, as I usually am, to kind of credit the shift in the polls to kind of broader macro trends, kind of normal midterm dynamics rather than individual campaigns. Okay, I want to come back to that. But Nate, I'm curious if you have a different take on this at all. I think the kind of macro level take that I have is that...

People tend to exaggerate the importance of advertising like the lay person does. Right. But it probably matters some at the margin, you know, maybe on the order of a percentage point or two. I think it sometimes matters more at the start of races than at the end. At the end, voters are pretty full and saturated of coverage in the early part of the campaign. You know, if one candidate in June of an election year is out with a

some big theme in their ads, right? When people are not paying attention, that theme can affect voter perceptions, can affect media coverage of a race potentially. You know, the other general thing is the further you go down the ballot, the less people know about the candidates. So the more advertising tends to matter, right? I tend not to think that you're spending your money very well on a presidential race in the House district or maybe something like a state Senate district. It can have larger effects.

Yeah, to your first point, Nate, about earlier in the race, I completely agree. I think an example of that this year is in the Ohio Senate race. Tim Ryan kind of went up on the air a lot earlier and more heavily than J.D. Vance did and was able to kind of define the race a little bit. And as a result, that race is closer than expected. In addition, I think like when you have like a huge advertising disparity, like one candidate spending millions,

millions and millions of dollars and the other candidate is spending nothing, that can matter. I think that's been the case in the Pennsylvania and Michigan governor's races where the Republicans only just recently went up on the air. And I think that probably does explain partially why Democrats up until recently, as Gayla noted at the top, were rated as solid Democratic in our forecasts, even though those are normally pretty swingy states.

Yeah, can I point out one other thing here that's a little complicated is that spending on advertising tends to be correlated with spending on other types of campaign activity.

So one kind of famous example of a case where campaign activity probably did move the needle was Barack Obama won the state of Indiana in 2008, which seems a little bit crazy when you think about how red Indiana typically is and has been since. But that was a case where the Obama campaign put –

Indiana on their radar and the McCain campaign did not, right? I think actually we're correct not to because it wasn't going to be a tipping point state. But yeah, so if one campaign campaigns in a state, so you're knocking on doors, you have yard signs, you have advertising collectively that might make an impact, but it's kind of hard to disaggregate the impact of advertising from other campaign activity. Esoteric point, but worth making. So we've established that

Democrats, at least in the final three weeks, seem to be spending more on these high profile Senate races. But we've also seemingly put high profile Senate races like Georgia, Nevada and Pennsylvania in the category that's probably saturated with ads. I'm sure listeners in the states can probably tell us themselves how saturated the market is. But when it comes to down ballot races like the House or even maybe Democrats,

races that have gotten a little bit less national attention, like gubernatorial races and even state legislative races. What does ad spending look like? Do Democrats still have the advantage there?

Yeah, Galen, in general, Democrats have had the kind of advertising edge up and down the ballot, although I guess I don't know about state legislatures because I think typically those don't get a lot of television advertisements, obviously, because districts are so small. And so it can be very inefficient, right? You know, normally, if you're doing a broadcast television ad, you're going to air it in like the entire market around, you know, Chicago or something like that. And so if a district is only kind of one sliver of that, that's not the best way to spend your money.

But in terms of the House, according to Ad Impact, we've seen over $1.2 billion in House spending this cycle. And Democrats do have a small edge over Republicans, $409 million to $352 million. In addition, on gubernatorial races, Pluribus News, which is a great new outlet that's kind of focused on state-level policy and elections,

They did an analysis recently that also shows that Democrats have outspent Republicans in nine out of the 11 races that they analyzed. Is it rare that you see the party that's lagging in the polls spending more money than the party that's leading in the polls?

No, not necessarily, Galen. There is a lot of kind of like trickiness with assessing the causality. Like I remember back in college when I was studying political science, one really interesting study was that kind of the relationship between ad spending and winning was kind of the opposite of what you would expect. Like often like losing candidates like spend more and like it can be because like if you're like in a close race, you're going to try to spend less.

more money, but it may not necessarily like you like you may be trying to overcome something like a disadvantage in other areas, but you aren't necessarily that isn't necessarily going to going to win out. So in general, I'd also kind of add that like these last few cycles and granted, they've been good cycles for Democrats 2018 and 2020. But we've seen Democrats raise a ton of money. They've been really good at that aspect of of politicking, and it hasn't always obviously translated into wins.

Okay, so to close this out, we've talked about maybe some circumstances in which ad spending does matter, some circumstances in which it doesn't. Is this all told a good or bad use of data? And I'll repeat the quote. One major ray of hope for Democrats is that the ground gained by Republicans in statewide polling of late corresponded quite sharply with a heavy TV reservation advantage. That is decidedly not the case in the last three weeks in the core battlegrounds.

Certainly in some midterms, you have the out party has a major advantage in enthusiasm and fundraising and everything like in 2010 or 2014. Right. We're not in that paradigm this time. I mean, clearly the GOP does have an enthusiasm advantage, but Democratic enthusiasm is also quite high, according to the latest polling. And the fact that the ads are fairly equal, if not higher.

if not favoring Democrats a bit, I think that's relevant, right? Especially talking about like, we clearly have seen movement toward the GOP since Labor Day. Will you have further movement toward the GOP in the last two weeks? In which case the Senate might go from a toss up to leaning GOP. I think the fact the advertising might not help them is a reason why you might expect maybe we're going to come into election day with this 50-50 Senate and not expect further movement, I guess.

Yeah, I mean, that's fair. You know, I think that to kind of the original point, you know, television ads don't matter a ton. I think they matter less than, you know, your average voter might assume, but they can matter on the margins. That's been the consensus of the research. And I think to that end,

Sure, if Democrats have this advantage in the final two weeks on the TV ads, it could kind of mitigate their losses. But I don't basically what I don't think is that Democrats should should view, you know, should see that, oh, you know, Republicans were airing a bunch of TV ads these last few weeks. And that's why the polls swung toward them. And then in the next two weeks, because Democrats are airing more TV ads, the polls are going to swing back toward them. I don't think that the correlation is that strong.

All right. Well, of course, we will see what happens. We're going to leave things there for now. Thank you so much for joining us today, Nathaniel. Thanks, Galen. Let's talk about that recent poll showing preferences this cycle amongst Latino voters.

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

As listeners well know, one of the most significant shifts in the electorate in 2020 was an eight-point swing toward Trump amongst Latino voters. So what are Latino voters thinking with two weeks to go until Election Day? We mentioned on Model Talk last week that obsessing over crosstabs in national polls of all voters can be a fool's errand. The samples of Latinos or other groups like independent women, as we discussed last week, in those crosstabs just aren't big enough. The answer is to oversample whatever group you're asking about.

Well, last week, Axios and Noticias Telemundo released a poll of 1,000 Latino adults, a sample big enough that the margin of error is under four points, as opposed to the double-digit margin of error we discussed last week.

In the poll, Latino adults preferred Democrats over Republicans by only 15 points. That compares with a 30-point margin or greater that Democrats usually see. But perhaps the most notable number is that about a quarter of Latino adults said they didn't know which candidate they'd support, and another 17% said they wouldn't vote. With us here to discuss, back on the pod, is Carlos Odeo of Equis Research. Welcome back, Carlos. Thanks for having me back.

So here's the question. Is a poll with this large of a group of undecided voters a useful poll two weeks away from Election Day? I say no. And I say with deep sadness because I do enjoy the poll for other reasons, but both in the way that the undecideds reported and also in terms of what the population of the poll was.

This just doesn't seem it, especially this close to an election. There are high numbers of Latino undecideds. You do see kind of late breakers. It has befuddled, you know, forecasting in like a Nevada in past cycles. But when I say large number of undecideds, I mean five to 10 percent. I don't mean 24 percent. So this does seem like very high. And again, for a few different reasons, good poll, but just not a poll designed to be

feed this conclusion. Okay, so let's break that down. One, why is this poll not designed to determine that conclusion? Well, first of all, it's a poll of national Latino adults. So already you're saying this is Latinos 18 and up. Some percent of Latino adults aren't going to be eligible. Significant percent are going to be either legal permanent residents or undocumented.

From there, of those who are eligible, some percent are going to be registered. And then, of course, if you want to start adding likely voter screens, some percent of the registered are likely voter universe. So already, when you look at the the generic ballot numbers here, you have 33 Democrat, 18 Democrat.

Republican, which looks very low and it is very low because some percent are saying they're not registered to vote or they're not eligible to vote. They didn't ask separately, are you eligible? Are you registered? So all in generic ballot question, you are trying to capture both who is eligible and who is voting and then of those who are voting, which way they would vote. And so you end up with this kind of confused,

very low number that doesn't give you a good picture of the population we care about here, which is at minimum registered voters. But it seemed like you were still interested in this poll as like a good use of polling to get some sense of how Latino Americans are thinking about politics, but maybe not a good sense of how they're going to vote or turn out in 2022. So what is the what is the like value here?

Yeah, for sure. And that's why I said with some sadness, I would say anything negative about it, because frankly, I can never get too upset about a Latino focused poll that's done very thoughtfully. Certainly won't get more upset about it than I do about breathless narratives from, you know, like based on tiny, unrepresentative subsamples and across that's.

Again, just isn't designed to support this conclusion, but it is designed to understand what are the kind of attitudes and issue preferences of Latinos kind of writ large within this country, many of whom aren't fully represented in the political process. And so, first of all, methodologically, it's like a dream. There's just.

you know, they're waiting on language proficiency. They're taking consideration specific Hispanic origin, you know, Cuban versus Puerto Rican versus Mexican or what have you. There's they're waiting on urbanicity. They're looking at 2020 weight recall the way they build their panel. You know, it's like a it's a proportional, you know, it's like Pew standards of proportional panel. And so it is.

in many ways, well-designed. And then they're asking, you're getting good insights on things like the fact that a majority of Latinos seem to feel that they don't have the same opportunities as other groups in this country. That's great for governance. That's really important stuff to know for governance, to understand what issues they think the next Congress should take up. That's really important stuff that you want to know for the actual running of this country.

We're here talking about a very limited interest, though, which is how are people going to vote in the upcoming election? And it's not designed for that. Carlos, I'm just curious, when it comes to those preferences for the sake of governing beyond just the horse race question, what are those preferences? It does seem to see you see the the the desire to see action on the economy, just like other like other members of the population. So Latinos are not

incredibly unique in that way. But you see the sensitivity around economic, you see the incredible pessimism around their economic projections going forward, all factors that are going to be important. You know, I think there's interesting nuances around how people are thinking about public safety, crime, guns that are all worth legislators taking into consideration. So Nate, looking at this poll, is this for your purposes a good or bad use of polling?

I think, well, I agree with Carlos's comments. I mean, I think it's like, in some ways, if you like want to get like an honest snapshot of all Latino adults, I think it's a very good use of polling, right? And like, similarly, if you took a poll of all American adults and you really made efforts to reach hard to reach groups, a lot of people are not registered to vote. A lot of people have no interest in voting. Some people aren't eligible to vote, right? Yeah.

A lot of people in a poll saying they're undecided and it's a poll of adults, that means they're not going to vote, right? It's another way of saying I won't vote is saying I haven't really thought about it. I don't have a preference. However, it is unusual this time of year when we're two weeks from the election, polls are usually taken among likely voters. You know, frankly, there's like a probably a bias in polls for people who are more politically engaged and are less likely to be undecided. But maybe that bias is OK because those people are the ones who turn out

to vote anyway. So it's like an honest snapshot of the overall electorate, I think. But it's kind of apples to oranges in terms of saying like, are Latino voters more likely to be undecided at this stage? I don't think you can conclude that without applying the same methodology to like, to all Americans, right? And also like, because, you know, a higher percentage of Latino voters are not registered to vote or not eligible to vote than like white

voters. That's not the same thing as necessarily being undecided if they are able to vote. So it just, you know, it's hard to make apples to apples comparisons here. So, Carlos, how would I know you've been doing some polling throughout this year? How would you compare the results of this poll amongst Latino adults in general to the polls that either you've conducted or that you've dug into that look at either registered voters or likely voters?

Yeah. And again, you're you are reading tea leaves somewhat here. Right. Because of what Nate is saying and the ways that truly is apples to oranges. That said, if you take the thirty three eighteen, you know, my friend Jonathan Robinson, a catalyst Democratic data firm would say if you're going to compare to past results, you have to you have to two way it. You have to take out the undecideds and the don't knows. Look at the Democratic vote in this case as a percent of just only those voting for the two major party candidates, literally Dem divided by Dem plus rep.

And if you were to two way 33 plus 33 to 18, you get something that's like 65 percent two way. Well, 65 percent two way is in line with 2020. It is just shy probably of 2018. It's kind of about in line with 2014. So 65 percent, I'd say, is fairly stable picture for Democrats, again, with all of the caveats of how you might interpret this poll. In our numbers, we're looking at generic ballot results.

probably in around 60 percent, 61 percent, two way, just 54, 35. If you look at it straight, a little higher for U.S. Senate, actually U.S. Senate races, Senate Democrats seem to be doing a little bit better among Latino voters. So I'd say that by that measure, this poll seems to be roughly in line with what you might expect to see in other recent surveys of Latinos that were intended to see the preferences of registered or likely voters.

One thing that I think is always worth mentioning is like, you know, we don't actually have any one definitive objective measure of how Latinos voted in 2020 or every other election. Right. The exit polls are prone to error in many of the same ways that regular polls are. Right. There are different estimates. You know, Catalyst, I think, had a different I think.

Yeah, I mean, I think the estimate's more realistic in part because, like, if you go to, like, the counties that were the most heavily Hispanic counties in 2020, they all showed a rather large shift from 2016, right? So Catalyst showing this large shift is more plausible than, like, the exit poll, which didn't show much of a shift. But exactly how much of a shift, we don't know exactly.

Yeah, exactly. And so when people say, for example, well, this Latino poll wasn't a weighted to 2020 vote, which 2020 vote benchmark would you like to weight it to? The reason you weight general polls of population is you have like a firm benchmark. You know what the results were among Latinos. You're picking among the options that are available. And we generally do use Catalyst for those purposes. But it's.

It's not a precise measure in the same way that literal votes coming from the secretary of state are. So the headline for this poll is Axios Ipsos Latino poll colon warning signs for Dems to re-ask the question. Like, is that a good use of polling in the sense that the takeaway from this poll is that things aren't good for Democrats amongst Latino voters? I don't think you can include that from this poll. Can you conclude that from other polls? I would say it's a, well, first of all, um,

I think you're going to see a mix of narratives when it comes to election night, when it comes to Latino voters. And I think there are warning signs in other polls for sure. I think most of them are showing stability, frankly. I think it's when you get down to, let's say, this level of, I don't know, Florida, where you do see something that looks more like a collapse. And everywhere else, you're looking for something that isn't a rebound from the levels that you saw in 2020 for Democrats.

is largely in line and actually big disparities even within a state. I think you're going to see a Mark Kelly and a Hobbs in the governor's race do somewhat differently among Latino voters. So it's also just not a standard. You're going to see some ticket splitting. There are a variety of factors here. I'd say the overall picture that in polling is there is still some uncertainty. So the undecideds are not, what is this, one in four?

But, you know, in our poll with pushing, I think you have something like 16 percent of Latino voters in battleground states who are undecided in at least one of the three big measures, governor, Senate or generic ballot. About 16 percent in an individual race. For example, the Senate race in Nevada, it's like 5 percent undecided. There is a big difference between 60 and 63 percent or so, but 5 percent decides a little bit of a different picture, but it's still meaningful if folks are going to break late in this contest.

And I'm curious, you mentioned that that polling result is with pushing, which is this idea that a lot of people, when you ask them, will just say, oh, I don't really know. Like a lot of people just aren't paying attention to the election across the entire electorate, regardless of what subgroup you're looking at. Without pushing, do you see similar undecided numbers to this poll? No, no. Yeah, I mean, pushing gets you two to four percent leaners, something along those lines. You're not seeing huge numbers.

movement of this variety. Yeah, I think to the point about like, you know, I mean, it's long been time for like a more sophisticated media narrative about Hispanic voters. I think in some ways it has gotten more sophisticated in recent years. But Latino voters have always been a group that is, you know, you would say cross-pressured. They have beliefs that might push them toward different directions. And I think the mistake that was made kind of pre-2016 was assuming that, oh, this is just a group that is like

solidly within the Democratic base was obviously not as a blanket assumption correct, right? Or to assume, oh, let's just kind of lump Latino voters and Asian voters and Black voters together under voters of color, right? Like that doesn't really work. They're all different groups and they have different preferences and beliefs. But if you look at like our state elasticity scores, which is how much does a state swing from election to election,

States with large Latino populations often are relatively swingy. So there's a difference between being undecided and between kind of being in the middle of the spectrum where you may split your ticket, you may vote for different parties in different elections, you may be open to persuasion, and it's not just a matter of turnout. That's a great way to put that. And I would say roughly, you know, we're looking at like one in four Latinos who are by some measure persuadable.

in this electorate right now. And then also when you have a high turnout election, what you're getting is a lot of voters who do not have firm partisan preferences. And I think there's there was this conventional wisdom that if every Latino voted, it would be the sky high Democratic margin. But actually, when you're looking at voters, we're kind of on the periphery of the electorate. And academia supports this sort of more broadly. You have people who just aren't

quite sure which party they belong to, aren't quite sure where they fit on the ideological spectrum even, and are going to be sensitive to movement in the particular race, candidates, and to the general environment. And this is where, you know, like the vibe theory of electoral politics is going to come in. But when you have a high turn on election and you have more peripheral voters coming in, to some extent, a lot of those voters are going to shift in the direction of the vibe. Very scientific. Yeah.

In some ways, is this Axios poll a good example of that? That when you're looking at all Latino adults as opposed to likely voters who are going to be more tuned into politics and have clearer partisan preferences, when you're looking at the sort of the whole population, it is a lot less clear that Democrats have that like 30, 32 point advantage.

Yeah, I think it's clear that you that there's a larger share of Latinos who actually are pretty more Democratic than Republican, but it's not a majority. It is actually a plurality of Latino voters who are not firmly set in their ways. They are more aligned with Democrats. Again, when you come back to their issue preferences, generally speaking, they are aligned with Democrats. They trust Democrats on Democrats.

or view Democrats, for example, as representing people like them more. That's one of the questions you had in this survey. They see Republicans as being more outright hostile to Latinos. So you have these voters who, as Nate said, are cross-pressured in some way, who maybe are open to persuasion right now because of their economic anxiety, are looking over at the Republican aisle and aren't exactly persuaded by what they are seeing over there.

Um, so aren't firmly Republican either. And they're just kind of in this limbo state in between. And I do think this poll does validate the sense that there is a lot of swing in the Latino vote, um, a swing that hasn't been fully capitalized on by either party. Yeah. Yeah. What are the, as we wrap up here, what are the races that you're sort of most curious about in terms of how much the Latino vote might swing or swing the election altogether? Yeah.

Obviously, everyone is looking at Arizona and Nevada. And I think that's that is right to do, because that's that's where they were. Both parties were really contesting to the max. Right. And so if you hear you have an example where both both parties understood the importance, both parties went in and are competing for the vote.

And does that end up in a stalemate or do Republicans make gains? If Republican make gains, then you can start telling a different narrative. If you're seeing something looks like stability, then I think it's a different conversation. That's there. Obviously, a lot of people want to talk about

South Texas. I do think the Mayra Flores race is interesting because frankly, Mayra Flores should lose the way that district is drawn up. She should lose that race. If she wins, I'm willing to like entertain a different conversation. But even there, South Texas is different. There's like a unique set of factors happening in South Texas, which by the way, is commands like 98% of the media attention, but it's 10% of the Hispanic vote in Texas. And South Texas and South Florida are both very unique.

Florida is going to be a cautionary tale for Democrats. I think you Democrats would be lucky to get 50 percent of the Hispanic vote in in Florida and would be happy to see it. And then on a third level, I think you have to monitor what happens in a state like Wisconsin, Georgia, Pennsylvania, where the Latino vote is like two to three percent, but matters a lot in these close races. And do you again see a picture of stability there?

or do in fact Republicans make gains? If I had to guess right now, I'd say it looks a little bit more like stability, but I think it'll be fairly telling once it's all said and done. Yeah, I was gonna say to that latter point, right? How do Latinos vote in states where, because that two to three is gonna become

four to five, five to seven, right, over time. So do Latinos in states that don't have a substantial Latino population vote differently than those that do, right? That will become a more important question electorally over time. All right. Well, once we have results, we'll have to talk about them with you, Carlos, but let's leave things there for now. Thanks so much for joining today. Thanks for having me. All right. Let's talk about what we mean when we talk about the educational divide.

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

If you've listened to this podcast for any length of time, you are familiar with the idea that voters' educational level has become a dividing line in American politics. The current trend dates back decades, but it really turned heads in 2016 when education was a better predictor than income in terms of how voters cast their ballots.

Since then, it's become one of the most common ways to describe our cleavages in America, including in this current election. But what does it actually mean? Why is a college education a dividing line? What types of beliefs and preferences underlie it? Senior politics writer Monica Potts addresses these questions in her first installment of our new Invisible Divides series on the FiveThirtyEight website. It's a great piece, so I encourage folks to go check it out. And also, Monica, welcome to the podcast. Hey, thanks for having me.

Let's begin with just the overarching question, Monica. Why is education a dividing line in American politics? And I know that you could probably write a whole thesis on this, but if you had to synthesize it as best as possible, what would you emphasize most? Well, I mean, I think that having a college degree versus not having a college degree is actually a real socioeconomic divide in American life as well. So it's kind of not surprising that it's also associated with political differences in

So, I mean, you do have, if you have a college degree, you're more likely to live in a city or a suburb. You're more likely to marry someone that also has a college degree. Over the course of your life, you are more likely to have a higher income level. So those are things that really do sort of set you apart from someone who doesn't, who's

more likely to live in a rural area, less likely or more likely to earn less over the course of their lives, those kinds of things. But I think the surprising thing is the kind of the depth and the of the political divides that are emerging in those two groups. Nate, how do you answer this question? Why a college degree just seems to matter so much? I think it's increasingly the way that class is determined in America. You know, partly for a long time, um,

Not enough people went to college for like college to get a voters to be that useful, like an organizing structure for a party. But you now have a party, the Democratic Party, that's kind of organized around in a lot of ways the needs and interests of voters who went to college.

Certainly a party where everyone who works in the party, this is also true of Republican elites in some ways, right? But like people who work in the media, people who work for the Democratic Party, obviously people in academia are people with a lot of academic credentials. But no, it is like, I don't know, like in some ways it kind of feels to me like why did this happen?

Take so long to be reflected electorally. Right. I mean, I think in a lot of areas, whether you went to college or not and what college you went to predicts a lot of attributes about social class and attitudes towards social issues and political issues. And so.

In some ways, Trump's great insight was to realize that actually a majority of voters did not go to college and maybe to be on that majority side, although Trump, of course, won a minority of the popular vote, but there was fertile electoral territory there. Yeah. So to maybe put some numbers to this, today in America, about 35% of people have a college degree. In the 1950s, to get at your point about

you know, not too long ago, not enough people had a college degree for it to make a difference. It was only about 4% of Americans who had, you know, a four-year college degree. But you started talking, Nate, about the ideas that underlie having a college degree, like what those beliefs are. Monica, which is part of what you dug into in your recent piece, and you even conducted a poll looking at sort of what those beliefs are. What did you find?

Yeah, I mean, well, first, just to say, too, only about a third of American adults have a college degree. But much more than that go to college and don't finish or leave with an associate's degree. But about 40% of Americans just never go to higher education at all after they graduate college right now. But, yeah, those are just higher numbers than there were a generation ago. And, you know, we started to ask people sort of not just –

Whether they had a college education, you know, but how they viewed college education. And so we found that about half of respondents, 51 percent, agreed that a college education is the best way to get ahead in the U.S.,

But when you looked at the partisan differences there, 71 of self-identified Democrats agreed, while 37 percent of self-identified Republicans agreed. So I mean, agreed. So whether it was like even worth it to go to college was a question that they didn't agree on. They were also asked a series of questions to sort of figure out how they felt about higher education's role in American society. And we asked sort of questions like.

College, do you agree with this statement? College makes you lose common sense. And 57% of respondents disagreed and 37% agreed. But of those who agreed, 65% plan to definitely vote Republican candidates in the midterms. So I actually think college is...

you know, bad that you go to college and you lose your common sense. A vast majority of Republicans agreed that colleges teach liberal propaganda, that people go to college and they are indoctrinated in liberal propaganda by their college professors. They increasingly think that those ideas are coming down into lower levels of education, like high school and even elementary school. So, so they, you know,

It's not just about sort of whether you go to college and get a better job, but also what's happening on those college campuses and whether it's good for American society, whether you pick up political ideas that they disagree with, those kinds of things. So I think there's been something of a debate about why, because a lot of what we're talking about here is,

whether voters have something of a liberal ideology, because that's the thing that accompanies oftentimes a college degree, and whether students arrive on campus with a sort of more liberal worldview or whether they leave campus with a more liberal worldview. The reading that I've done on this is that it's both, basically, that both

liberal people are more likely to end up in college and college is likely to make you more liberal. Is that sort of, is that backed up by the research that both of you have done? Like one thing I'd say is that, um, because educational polarization is a relatively new phenomenon, I think inherently the jury is still out on some of these questions. Right. And also the kind of style of education in college is changing. Like basically like, um,

like college professors as a group have always been democrat liberal left-wing however you want to describe them um but maybe not in quite the extreme numbers you have now where in some professions it might be you know 20 to 1 or something people who are registered democrats versus republicans or things like that and so um so that might be different i i don't know that like kind of

Well-educated elites are used to the fact that like now they kind of all do have the same political preferences. I'm not sure they're adjusting for that enough and that they might think that they are.

objective experts, but like any people are subject to partisan pressures. And I mean, you know, I don't know how much I want to get into a discussion of COVID, but I thought COVID was a case where the expertise started and stopped and where partisan political views that were pro-democratic, anti-Trump started and stopped. That was a big, messy conversation in COVID. And I think in ways that wouldn't have been as true

10 or 20 years ago when more public health professionals and epidemiologists were, you know, Mitt Romney voters or Bush voters or something like that. I think it's all kind of new.

Yeah, I would just also say I don't think it's new at all for college campuses to be places where ideological young people sort of push the boundaries on what society expects of them and how they want to interact with society. Or that, as Nate said, there's probably an imbalance in how liberal college professors are. It's just part of the nature of the academy. But what is new, I think, as well is that

I think a lot more of these conversations are extending into the public sphere via social media. And so you see people having these conversations on Twitter or to a lesser extent on Facebook and other sorts of places where people who aren't on campuses can see them. And maybe people who are like non-traditional college students can see them and they don't comport with their realities. So I think that there's just more kind of fodder for the idea that these are just –

really liberal places where people are doing things that people don't agree with. So yeah, that's part of it too. Yeah. That, that, that Twitter thing is because it's like, I mean, the image of academia that you would get from Twitter is that a lot of these people are batshit crazy, a, an extremely left-wing B and,

But, you know, I have friends who teach at colleges and universities and I interview a lot of people for my book. A lot of the people are at universities and a few times a year I'll go and do some type of talk on a university campus. And like my dad's an academic and like I'm not sure that quite describes the reality, the Twitter sphere. Right. Yeah.

Even at elite colleges, I'm not sure it describes it. And it certainly doesn't describe it at non-elite colleges. Well, sure, but that's honestly not even really the question here because we're talking about Americans on the whole, big political divides, not just like Twitter is like, yes, you're right. It's small enough to the point where like,

why get into it for the sake of this conversation? Like we're talking about, for example, in, you know, 2020, we saw that college graduates, according to the exit polls, and those caveats, they're made up 40% of the electorate, 55% voted for Biden, 43% voted for Trump, non college degree, 48% voted for Biden, 50% voted for Trump. Of course, if you break that down by race or ethnicity, you get significantly different numbers. So you know,

The number that draws a lot of attention is white voters without a college degree. They made up 35% of the electorate. 67% went for Trump. 32% went for Biden. So this gets beyond, yes, whatever is happening on Twitter, the vast majority of Americans aren't paying attention to it anyway. But like, I'm curious for the sake of this.

And why that, you know, more than even income, as we mentioned, is driving behavior. Because I don't think it's, you can say like, oh, people are looking at conversations on Twitter and thinking, wow, that's nuts. Like, I'm going to vote this way or that way. No, I'm going to push back on that a little bit because I do think Twitter is important to like elite discourse. And it's important to like how the media, it's very important.

how the media covers campaigns, right? I think the crazy professors who are tweeting up a storm can have fairly big impacts on the perceptions of academia and public education. And they can certainly influence the Democratic Party's strategic choices. Let's look at one policy choice, or a couple actually, right? Forgiving student loan debt

is very much something a party does if it says we are trying to protect people who are college educated voters, right? Although some people didn't finish college, but like that is very much in kind of the class interest of this new kind of class of Democrats. Um,

The lockdowns that Democrats tended to favor under COVID, although they're not lockdowns in the same sense of like China or Australia or something, but like work from home policy tends to help people with white collar jobs who went to college. The way different messages are used in advertising, the language that is used tends to be crafted by very left wing kind of college educated people. So it has a lot of influence on the way the parties are perceived. Right.

Yeah, I agree that when things happen on Twitter, they get picked up and they reverberate in ways that people who aren't on Twitter still see them. And so if there's a story about something that happens on a college campus, it's going to be on Fox News and it's going to be written about by the New York Times because –

you know, it's just part of what they consume as well. And so then people hear about it or see it in a way that's like filtered through these perceptions that they already have of Democrats being elite and being liberal and look how liberal they've gotten those kinds of stories. And so I think it has a bitter impact that bigger impact than just who's on Twitter. Yeah, to get at some of those themes you wrote in your story on this.

The division is about more than simply getting a college education or having a higher income. College and relatedly elitism are concepts that seem to be linked to otherwise unrelated ideas, like support for LGBTQ plus rights or racial justice, which have in turn become associated with the identity of the Democratic Party. Increasingly, it seems that when politicians like Trump rail against the elite, a word that might otherwise mean a level of wealth and power he himself has, it's meant as a cultural designation, not a socioeconomic one.

Can you talk a little bit more about what you mean there and how that relates to, you know, the diploma divide? Yeah, I mean, I think you really saw it with the student loan forgiveness debate. You saw a bunch of politicians like Tate Reeves, the governor of Mississippi, Mitch McConnell in the Senate tweeting about how, you know,

And President Biden's student loan forgiveness plan was a giveaway to the elites, to these high earners that took on too much debt for stupid degrees that had no place in the labor market, these kinds of ideas, and that this was just a giveaway to everyone who's elite and a snub to hardworking Americans who earn a paycheck, you know, who get paid by the hour.

And you kind of see that. And what's interesting about that is because, you know, I mean, the Biden administration says that student loan forgiveness is largely going to benefit people who earn less than $75,000 a year. As I said, there are a lot of people who go to college who take out student loans who don't finish and so they don't get the income benefits that you get from a bachelor's degree over the course of your life.

So it's a much more complicated picture, but it feeds into this idea that like there are just these elite people goofing around on college campuses for four years studying like, you know, women's history or whatever. So it's all these cultural things that are tied in with what people think liberals believe and whatever.

You know, it's kind of pinging on a cultural level as well as an economic level. So like having a college degree or not having a college degree almost becomes a cultural tribe. Like you associate with what that means beyond just like, oh, OK, I spent four years on a college campus.

Yeah, and what you believe that you were indoctrinated by these liberal professors who told you all the things that people who didn't go to college don't believe. One other thing to point out, I think people make this assumption that, oh, inevitably more and more jobs will require a college degree and so eventually –

Everyone goes to college. The number of people going to college has actually started to flatline or even decline a little bit. So it isn't necessarily the case in a quote unquote modern economy that there is demand for college educated jobs. We also see a big.

gender imbalance and who goes to, particularly who graduates from college. It's much more women than men now, which is interesting. So I think you may actually see going forward for people for whom going to college is a choice you could make either way as a marginal decision, that it may be selecting more unpolitical characteristics, right? That now people who are more liberal think this is the place for me. And now people who aren't will want to go into labor force right away.

New York Magazine, interestingly enough, also published a piece looking at this divide recently. And they try to get back to the root of all of it. Where did this begin? And...

They talk about the baby boomers as being the first generation that had enough people going to college that it sort of shaped their culture and beliefs. And there was this whole question as the baby boomers became adults about like, why are they so liberal? You know, of course, like during the 60s, how did this all happen? And one of the theories, I'm curious what you all think of this, is that.

The post-war era was one of the first times in American life where there were middle class masses who didn't actually have material concerns as children. That prior to this, lots of children, the majority of children in America grew up knowing what hunger looked like, etc., etc., and that economic concerns were absolutely chief for the vast majority of Americans.

But when you have all this wealth accumulation and a lot of kids growing up middle class and then going to college, their concerns, their policy priorities move from material concerns to post-material concerns, which is things like rights, like LGBTQ rights, feminism, civil rights, things like that, that sort of are in the category of post-material as opposed to...

And I'm sure that some people would argue with the designation as post-material as opposed to like, I just want to make sure that I can get a job, period, and feed my family, etc. And that this post-material versus material divide in terms of policy preferences is really the start of, you know, what we see today as a main dividing line in American politics. Does that pass the smell test? Does that seem plausible? Yeah, I mean, I think there are actually different

kind of synonymous versions of this theory, right? Sometimes on the right, you see the term like professional managerial class invoked in somewhat a pejorative way. But the idea that there are now more people with college degrees than necessarily have the ability to have highly influential jobs. And so they tend to debate ideas. They tend to compete for status by how well they adhere to

social norms of kind of the liberal social class. I mean, sometimes the term political correctness or wokeness is used in this context. But, you know, yeah, I mean, I think, I mean, look, if you go on

Twitter these days, you will see a lot of... Back to Twitter. Well, but this tells you how... I mean, we're talking about... I don't think you can just separate this out from intellectual elites, which as we were pointing out earlier, are different than economic elites, right? But there are people kind of saying, oh my gosh, how can these Republican voters vote based on inflation when democracy...

is at risk, right? Or when women's rights are at risk. I mean, that's a very professional managerial class attitude and saying like, actually, these people are cringy for voting based on pocketbook issues, right? That's that theory kind of put into practice. Yeah.

I mean, I think I feel like at some point everybody who writes about this became a really like a material determinist as far as culture goes. And I mean, I think it's part of it. There was pretty widespread affluence in the middle part of the last century for largely white Americans. There were lots of Americans who were left out of that. I think that.

That probably contributes to some degree. There were a lot of changes since then, though. There's been a decline in union density for working class jobs in the years since that period. There's been a lot of pushes for, I mean, people cared about equality before they came

had settled their material needs. So I don't know if it's that deterministic. But the forces at work after World War II, which was really devastating in the way people lived their everyday lives, all played a role in how American life reshaped itself in the years following. So yeah, and I think you could see some of the shifts date back to then, but it's all changing all the time. So I wouldn't put a pin in it that directly.

One of the things that you looked at as part of your piece, Monica, is how this divide could shape policy going forward for the two parties. What'd you find? Well, I mean, you know, one of the things that I think we've been operating under in the past generation, at least, is that college is a good thing for people to have access to. So if you are from a family where you don't have access to college, you should have access to ways to help you afford to pay for it. That

It's good for American society if people spend more time learning things in their education, if more education is better. It's just kind of been the operating assumption. That's not entirely true. As Nate said, there are a lot of kind of caveats in there about how we view the role of education and education.

people's individual lives and whether it's worth it for everyone and what it means for education to be worth it and how the labor market is kind of reshaping itself, especially post-COVID. But I do wonder whether national policy will change on some of those questions if, you know, we don't think it's important for everyone to have access to a college education or if we don't think it's important for everyone to go educate, change education on lower levels.

Yeah, I mean, one direct consequence could be, you know, in general state

colleges and universities are popular institutions. You have a lot of graduates from the big public universities. They have football and basketball teams that are popular. There was kind of a movement in kind of the Scott Walker era in Wisconsin a few years ago to try to reduce funding for public education. We may see more of a wave of that now that if college and school voters are seen as like kind of an interest group of the Democratic Party or maybe kind of universities in particular, there might be more

GOP efforts to reform the tenure system, obviously like Ron DeSantis in Florida has done some of that, or to reduce funding for public education in general. It might be a more tenable position now in the GOP. But also, look, I worry about what happens to the country when

the expert class all votes for one political party for two reasons. One is I'm not sure if one party is going to win elections half the time. I'm not sure how I feel about there not being very many experts in the party. And secondly, I worry about if one party does have all the experts, they act in partisan ways. I mean, this could be a long conversation, but I think like

experts acted in flagrantly partisan ways in the COVID crisis, for example, in some ways, although it was very difficult to solve some of these issues. And I wonder, you have underperforming experts on the one hand and no experts on the other party. That seems like it's kind of a recipe for

for disaster and kind of maybe self-perpetuating problems too, where if Republicans take steps to undermine academic credentials and Democratic parties make steps to try to serve them as an interest group, like I think kind of was done with the college loan repayment program, that seems like it becomes an accelerant for a lot of unfortunate trends in America today.

Yeah, I think, yeah, one of the big things is sort of decline in trust and expertise where, you know, when you have issues like the pandemic, it's going to actually really affect lives, for sure. I also anecdotally, I think you're going to, I do see people declining support for public education as an institution, even for their kids, you know, in states like Arkansas and Mississippi, obviously live in Arkansas, you're going to see people start to

question K through 12 public education's worth more and more and what that means. And so, yeah, I mean, that's just going to remake American life if you, I mean, that's a really longstanding American institution, free public education for K through 12. All right. Well, I'm sure there's so much more we could say about this divide and its consequences in American life and politics, but we're going to leave it there for now. So thank you, Monica and me. Thank you. Thank you.

My name is Galen Druk. Emily Vanesky is in the control room. Chadwick Matlin is our editorial director. And Ben Schelfifer is on video editing. You can get in touch by emailing us at podcast.538.com. You can also, of course, tweet at us with any questions or comments. If you're a fan of the show, leave us a rating or review in the Apple Podcast Store or tell someone about us. Thanks for listening, and we will see you soon.