The Poll Miner

Prefer to read? Check out the Audio Transcript

John Gerzema: What basically happens on these questions now is that they’re geared toward being anti incumbent. In this month’s Harvard Harris poll, 62 percent of Americans say that the country is on the wrong track. But if I had shown you that data four years back, it would be nearly identical, but it would be flipped.

Zachary Karabell: What could go right? I’m Zachary Karabell, the founder of The Progress Network, joined as always by Emma Varvaloucas, the executive director of yes, you’ve got it, The Progress Network. And this is our weekly podcast where we look at what could go right as opposed to what could go wrong. We try to look at what’s going well in the world, or at least we try to look at people that are taking a more nuanced, sophisticated, calmer, saner, less outraged, less fearful, less hysterical, less partisan, less political, less dystopic, less dystopian view of what is going on in the world. And one of the ways that we all assess what’s going on collectively is through this very modern, modern as in, in the past century, institution known as polls, where a group of people go around and ask another group of people what they think of X and polling has become one of the tools that we use to try to figure out whatever we are, as opposed to whatever I am.

So we’re going to talk today to someone who I think probably does a better job, given that much of that polling, or at least a portion of that polling in a political election year, is driven entirely by horse race mentality, who’s going to win, who’s not, who’s going to get the most votes, who isn’t. So Emma, who are we going to talk to today?

Emma Varvaloucas: So today we’re talking to John Gerzema. He’s the CEO of the Harris Poll and he’s a pioneer in the use of data to identify a particular social change. A Harris Poll, you know, does a lot of different kinds of polling, in particular, helping companies anticipate new trends and demands, but also lots of general polling about Americans, their mood, what they care about, what they’re looking at.

He has read for numerous different publications and is also the author of a couple books, among them The Athena Doctrine. So let’s go see what John has to say about polls.

Zachary Karabell: Let’s do it. Welcome John, to our podcast. Welcome to What Could Go Right?. I thought it would be interesting. This is a spontaneous thought and therefore may be much less interesting in retrospect than it feels pungently in the moment.

But tell us about the history of the Harris poll, like, like, how did it come about? Who created it? Who was Harris? What was his goals? Was it different than Gallup or Pew or anything else?

John Gerzema: Oh, that’s an excellent question. You’d have to go back 60 years, but Lou Harris started the company up in Rochester, New York.

And if you imagine that time, these were the giants of industry. You had the Harris poll, you had Kodak. You had a few other companies of that ilk, they were doing some really interesting things upstate, but he had this sort of simple belief that the public’s voice mattered in leadership and that leaders needed to be held to account.

But his sort of mission was to try to bring the public’s voice into the debate on politics and then later into business. So he was John Kennedy’s pollster. That’s sort of how he got started and built a really fascinating polling company, that by the time of his death in 2016, Harris sadly sat inside Nielsen, and it wasn’t even the Harris poll anymore.

They’d taken the name down and they In their wisdom, they called it Nielsen Consumer Insights. We’re able to acquire the company as part of our holding company, which is called Stagwell Group. And the Harris poll kind of came roaring back. So we’re proud of the work we do.

Emma Varvaloucas: So I feel like with polls, obviously, what they tell us is brought into the public conversation a lot, but I guess until 2016, I don’t remember there being like public conversation about polls themselves, right?

Like there is such a downfall of the polls moment then to the point where people, you know, began asking, like, do these mean anything? What’s the point? So I wanted to give you an opportunity to give a full throated defense of polls.

John Gerzema: Got it. So that’s good. That’s my job. Yeah. So the 2016 election was probably the Janet Jackson, Justin Timberlake Super Bowl episode, great reference, you know, where everything that could go wrong, went wrong.

There were a myriad of reasons in retrospect that have been widely reported, but I think the real important thing to think about with “When polling goes bad,” is there are a lot of really bad polling firms. And you can see, even if you fast forward to the, well, in that instance in 2016, you had obviously Trump voters wildly under measured, right?

They didn’t weight their samples right. They made simple mistakes, like not understanding that many Trump voters, people in rural parts of the country weren’t online and they weren’t available for online polling. So simple things like that. Then there was obviously response bias, and there’s a lot of different biases that go into polling, but a respondent bias could be one, all kinds of different biases, but one of them is likability.

And so that would create shy Trump voters. One of the things we’re looking at in the future for this election is not shy Trump voters necessarily, but maybe shy Biden voters. So people that have not really stated their opinions, but might show up. So, polling is incredibly difficult, but the people that get it right, we like to think we’re part of that.

We were very accurate on the election last time, and as were many other pollsters. But you have to have the right resources. You can’t wing it, to be honest. It’s not a good sport to be in.

Zachary Karabell: So you mentioned that Lou Harris initially founded the company and was Kennedy’s pollster. And I have to admit, I’ve always been a little confused at the party affiliation part of pollster organizations.

So you see like Stanley Greenberg was, you know, Clinton was a Democratic pollster. And then, you know, Frank Luntz was a Republican pollster. And I would think that from a political calculus, there’s zero partisan gain from having a partisan read of what people might do in the future. Meaning, isn’t everybody interested in actually figuring out where people’s ideas are and then either trying to shape that or, or gain that accordingly?

So the whole Republican versus Democrat poll, I do find confusing. I don’t find it confusing of if you want to have a data point for an opinion columnist, So you want your poll to come to a foregone conclusion, you know, X percentage of Americans are pro-life, so therefore we should be pro-life or pro abortion.

But I don’t understand that part of it, right? Because you kind of want to know what people are thinking, regardless of which side you’re on and act accordingly, don’t you?

John Gerzema: Yeah. So the cynical side of that is there are many left leaning or right leaning political firms that will make a lot of money putting out their own polls.

They’re going to whip up their constituents, create fundraising, create political action committees, and there’s business in there and advantage. But we don’t look at it that way with my folks. I’ve spent most of my life out of New York. I was born in Montana. I lived in the Midwest for half my life. We try to get a very well representative group of researchers that sort of represent the country. In fact, we’re highly virtual. So we’ve got people all over the country and we just try to encourage them to be umpires and see both things. Cause that’s, what’s exciting, right? You’ve got to be able to look at any type of an issue and be able to argue it from both sides.

News Clip: How confident are you that the polls are reflecting the reality of, um, what’s going to happen a few months from now, even? Well, polls are, first of all, a snapshot in time as of today. They’re not actually the best predictor tools. Uh, it takes an analysis of what’s going to happen, what you think’s going to happen, but the dual nature of what voters think is really what makes American democracy so effective and, and what makes elections so unpredictable.

Emma Varvaloucas: So John, I’m wondering if you could, maybe the SparkNotes version of this, because I imagine there’s a lot of weeds to get into, but tell us like what makes for a good poll? Because a lot of the time I’ll see polls online and people will say, Oh, it wasn’t a good poll. It had a sample size of X. And I’m like, okay, but what is a good sample size?

I didn’t go to university for this; someone’s gotta at least give us the basics.

Zachary Karabell: A good sample size is everyone. Everyone is a good sample size.

You should poll all of us all the time with a hundred percent response rate.

John Gerzema: You gotta know it. Look, there are lots of things you kind of have to avoid. There’s things, phenomenons that are called river sampling, which is, the way that you collect your sample is related to different sorts of online promotions and motivations to get people to become respondents. So we work with a lot of different panels. They have really sophisticated fraud protection. So you’re just looking for things like reaching people where they are.

So in some instances, they don’t have broadband access. You’ve got to be able to weight your samples so that you’re getting the right people that represent the total population of the country. We do typically like 2000. It’s sort of a base size for us to be statistically accurate and that’s sort of in line with most notable pollsters’ research standards.

But I think, you know, the more interesting question in all this is Zachary, kind of your point. How do you write an interesting question, right? Because you can write questions, we call them carrots or ice cream questions, you know, so I can say, you know, Zachary, do you want like carrots or ice cream? And most people in the country are going to say, I’ll take ice cream, right?

So whenever I see anything that’s got 90 percent agreement in this country, you know that it’s not a good question. It doesn’t have tension. It doesn’t have trade offs. And it doesn’t go deeper to sort of find insights. The kind of stuff we’re proud of, it’s just things that shed light on culture. And one of the ones I was really proud of recently, we actually uncovered that grandmothers, that grannies were kind of the hidden engine of the economic recovery.

And the way we figured that out. was we found that two thirds of working parents that were holding their jobs said that they relied on a grandparent for caregiving, right? And 20 percent of those working parents said they would have lost their job without that support. So suddenly you just saw this really interesting thing that I don’t think I’d have really thought of, is like how important the support structures are for working parents, especially working women.

in America and the role that grandmas play. So that’s what we’re trying to do. You just try to go seek out the kind of untold stories and think like a journalist when you’re kind of going after these types of polling questions.

Zachary Karabell: Did you call the poll the granny economy?

John Gerzema: Dang it, we should have. I can’t remember.

We gave it to Fortune and they wrote something clever about it. That’s where my talents stop as a researcher, so.

Zachary Karabell: So on this thing, so let’s say you do a poll, you define your 2000 set, geographically, demographically, race, gender, urban, rural, etc. And you get a perfectly worded question. But doesn’t then the problem become in today’s particular world, I am sure I am not alone in that if I got a number these days from a number, an exchange I don’t recognize, I’m highly unlikely to pick up the phone. I doubt I would call you back if you left me a message. Those people who do online samples, which I guess for a while was morning consults or others, it’s selective as in, are you interested proactively on your own, unpolled, are you interested in pushing your information out there?

And that’s a whole other question. Who will do that voluntarily? I don’t have a landline or I do have a landline, but I think it goes automatically to my cell. So basically, how do you reach people and how do you get them to answer the phone in order to even do a survey today?

John Gerzema: Yeah, so we’re not doing telephone polling because you’re right.

Nobody picks that up. We haven’t done that for ages. I mean, there’s some very rare instances where we’ll still use the phone. We’re online, but we’re working with panel providers. And so what we’re doing is we’re actually getting access to fresh respondents that have been brought in to do a poll around a specific topic.

They would have to have knowledge of the topic. They’d have to have knowledge of the brand. I should say also, by the way, we’re not a heavy duty political firm. We have a Harvard Harris poll that we do every month to sort of track the mood of the country. But most of this work is done for brands, for marketing.

So to try to get people’s opinions on various topics. So no, it’s a hundred percent online.

Zachary Karabell: Do you do the Harvard Harris with the same methodology?

John Gerzema: Harvard Harris is done with the Harvard Harris poll through our political firm in D. C. Yeah, called Harris X. Meaning you don’t do phone. It’s still online.

It’s still, it’s online. Yeah. And the difference in it is it’s American voters. It’s not Americans at large. When you get to American voters, you find that they’re much more engaged in politics and they’ve got stronger opinions than the rest of us here in the country.

Zachary Karabell: And just one more on that. Do you worry about if brands are providing panels for pay that, I guess the pay doesn’t really matter, right?

I mean, it’s a nominal amount for people’s time, legitimate exchange of sort of time for a, I mean, these are modest side panels, right? It’s usually a hundred bucks, 150, 200 bucks. It’s interesting that we don’t do that for politics, right? It’s just something that we don’t do, right? We don’t think it should be done.

We don’t think you should pay for political opinion. Why, do you know why that is or how that came to be?

John Gerzema: Oh, I think it’s just the desire to want to keep the political process and the process of trying to understand voter opinions, you know, without any sort of commercial incentive. Again, as I say, like the majority of my work is all done in corporate business side and people are quite happy to offer their opinions.

What you’ve got to do is be very careful that you’re not getting professional respondents, people that are panel providers bringing back in over time. You’ve got to watch out for all kinds of sophisticated fraud. So there’s a time delay in answering an online question. You can see if somebody’s just going through and ticking a box versus someone that’s really thinking through, and you statistically measure those averages.

Try to make sure that the actual question’s been thought through and answered based on the style of question. A lot of different things sort of go into it, and then you end up with a margin of error, 2.5%. The important thing is, is you ask those questions lots of different ways and you go back and you maybe keep trending a question. One of the things that guys we’ve done since the beginning of COVID was just measure basic questions every week in America about America.

And so we kind of call it our America This Week report that we put out every week on LinkedIn. But it’s just looking at questions like, I feel secure in my job, or I feel my health isn’t at risk, or I feel the economy is trending in the right direction. And so that’s the easiest way to try to understand if you’re sort of in the ballpark on your questions or if you’ve got anomalies.

Emma Varvaloucas: So, given that you have these weekly polls, what is the mood of America right now? Because I keep hearing the same stuff over and over again, that people are pissed off about inflation and everyone feels that it’s a fractured society, but maybe that’s what you’re hearing too, but I would love to hear some other stuff.

John Gerzema: Well, yeah, I mean, I got two great stats for you. So this is from last week, 69 percent of Americans say the current chaos happening today makes me optimistic something might change and 73 percent of Americans say they are concerned that nothing will change. So, um, there’s your, there’s your answer. No, seriously, it’s a real interesting time and I could give you this.

This is just my chart. It looks like an EKG, right? This is our weekly tracking sort of different things in what we call the poly crisis. And I think what’s fascinating about today’s times is this has been a period in American history that we’re living through that’s sort of not unlike the depression, right?

Where you had four sustained years or World War II, where there was just been four straight years of just new crisis sort of coming one after the other, and obviously it went from COVID to the economy, to the banking crisis, to Ukraine, to solvency of US banks. It’s just seems like there’s something that’s coming sort of every week.

And I think that’s important to understand right now. The big optimistic side I kind of see on this is that I’m with the first stat that something’s going to change because things have been so bad and so sort of down for so long. This is a time of reappraisal and we see it in our data. Seven in 10 Americans believe society needs a complete overhaul to make significant change.

I’ve got all kinds of data about Gen Z. We call them GenZilla, but we think GenZ is going to move into an activist sort of position in the coming years based on the financial situations they find themselves in and the war that they have with old people. I mean, 71 percent told us, last week that they think older generations are short term thinkers exploiting next generation’s future.

Obviously only 31 percent of boomers agreed with that.

Zachary Karabell: That’s a good line. I mean, you know, the 69 percent who, as you say, think that things are so bad that therefore they believe that will create change for the better, it’s a little like Herb Stein’s old adage of something cannot go on forever, it will stop.

Right. So that. It’s the, if it can’t continue, it won’t, which almost seems like a syllogism and yet has a certain truth to it. I do wonder about this, and I don’t know if you’ve asked any of these questions at a granular level, just trying to get people’s attitudes about if confronted with data or statistics, that suggests that their views or beliefs are at least disconnected from the data.

It doesn’t mean that therefore they’re wrong. Do people ever seemingly reconsider their views or beliefs or, or no?

John Gerzema: Rarely do they reconsider their beliefs, but they’re happy to sort of live in their own opinions. We’re seeing that right now with immigration. It’s the largest single biggest issue in the country as of this past month.

And a lot of that, if you actually ask people, Have you witnessed the immigration problem where you live, you’re going to get far fewer people that can actually really say that with certainty with observation. But it’s like all other types of perceptions that exist in American culture, perceptions of crime, perceptions of definitely immigration, those things just become sort of self fulfilling prophecies.

We did a poll with the Guardian last summer when Biden was sort of at his lowest, not getting credit for anything. And everybody was so completely dour. I guess it was in the fall. A bunch of questions that basically every American flunked the quiz to. We asked questions like, is the economy shrinking or growing?

And the majority of Americans thought it was shrinking when it wasn’t. We asked if the GDP was up or down, which again, they got it wrong. We asked if the S and P was up or down and they had to filter to know the S and P. And then we asked about inflation. We asked that one really pointedly, we said, is unemployment at record highs or record lows?

And 62 percent said it’s at record highs. So it just shows you facts are facts. And that’s probably why my favorite polling fact is that 52 percent of voters say they rarely believe political poll results. So you’re dealing with a difficult business. But what becomes more important than sort of the anomalies is the underlying questions of why, like why people are thinking the way they are. And I think that’s the start of where things get really interesting. Like I said about this GenZilla, we started to weave and see a theme of activism in a lot of the things that young people were telling us. Feeling that older people won’t get out of the way, feeling that capitalism isn’t working for them, feeling that they’re starting behind financially, you know, each and every poll, all these different projects, you start to see these patterns and that starts to become the thing that I think is more meaningful.

Emma Varvaloucas: So I’m curious to hear if you want to elucidate some of the why’s for us on that, that you see as far as Gen Z. I do have kind of a halfway question before that, which is going back to our conversation about how poll questions are phrased something like whatever percentage of it was of Gen Z that think the older generations are short term thinkers.

Like I’m thinking about how I would answer that question, right? Like I don’t walk around thinking that the older generations are short term thinkers, but if someone put that question in front of me in a poll and was like, do you think that they are? I’d be like, eh, I don’t know. Yeah, probably maybe climate change, I guess, you know, that would be the thinking, but it doesn’t necessarily mean that I have like a really strong belief about that.

So yeah. Thoughts on that.

John Gerzema: Yeah. So. Again, it’s just looking at the patterns. First of all, just not asking leading questions, right? And again, trying to ask questions that involve trade offs, personal trade offs, you know, so it’s like, would you want more of something and would you give up something in return?

And so I guess that becomes a really important thing. But, we also have to try to go find when people are sort of not really asking the questions the right way. And a great example of that is we just asked this real simple question because there’s all this political toxicity right now. So we just asked Americans, In your opinion, which of the following words are more likely to bring people together or divide them?

And right off the bat, DEI was alienating by 38 percent of Americans, including 47 percent of Republicans. Now Only 27 percent of Democrats, it was like a 20 percent split. In short, it’s just become sort of a politically toxic word. Well, we went back and we started talking and we work with the Black Economic Alliance.

And we thought that the way that question’s asked is highly simplistic, right? And what we figured out is that DEI, like ESG, these have just become dog whistles, right? When you hear those, you actually don’t think like a human being. You think like a voter and you kind go into your own political ideology around these words because you’ve sort of been trained that these are bad words or good words.

We just went and asked the questions in an entirely different way. And we asked Americans around corporate diversity and whether that was a good thing or a bad thing as it related to them. Did they think that corporations would be better at customer service if they were diverse? They think they’d be more profitable for shareholders.

Do they think that reflecting the diversity of American population was going to make for better products and better services? And we have high support for corporate DEI. They actually think it makes stronger companies, but the questions aren’t asked that way, right? When the news cycle hits, it’s like DEI is a bad term.

So everything is, it’s got nuance in it.

Zachary Karabell: The lack of zero sum questions is a really good one as a way of getting about it. I mean, one of my bugaboos is knee jerk anti GMO reactions where people don’t entirely know what it is that they are opposing when they’re opposing GMO, as in like what that really means in a fundamental sense, even though huge majorities in certainly Western Europe and urban America would be against GMO.

But if they were asked the question, Would you support the growing and the isolation of certain plant strains that are far less water intensive and don’t require insecticides or pesticides, you’d have a very different response, right? It’s kind of the same question. One of my favorite zero sum ones, right, which is the famous 2015 poll where something like 35 percent or 30 something percent of presumptive Trump voters were in favor of bombing Agrabah, which is the name of the kingdom in Aladdin, a fictional kingdom in Aladdin, This is not apocryphal.

I mean, you remember this is, this actually was a polling question cause they wanted to get at attitudes. But of course, when asked, you know, do you want to bomb a country that you think may be a vague threat? Sounds like a threat without any trade offs, right? Would you send your children to die to prevent Agrabah from doing whatever?

It’s a very different question then. Exactly. Do you support the bombing of Agrabah?

John Gerzema: That’s classic. That’s great.

News Clip: Public policy polling got 30 percent of Republicans and 19 percent of Democrats to say they support bombing, well, Agrabah. Yes, by you millennials, that Agrabah. Home to Princess Jasmine and the street rat, Aladdin.

After the news broke, headlines largely focused on the Republican respondents in the poll, which kind of misses the sizable percentage of the other side of the aisle.

Zachary Karabell: In corporate land, separate from, do you like strawberry vanilla YooHoo versus more caffeinated Mountain Dew, do you find that there’s sort of company data about consumer behavior that would be more useful if it were far more in the public realm?

Because really a lot of companies want to… they want their product data and their corporate data to be their own.

John Gerzema: Yeah, go a bit deeper on that. What are you thinking?

Zachary Karabell: Company X is finding something vital about their consumer base. Could that information have constructive public good applications, rather than it simply being kept by a company for its own ability to sell its particular product?

John Gerzema: A lot of what is changing today is the kind of new style corporate advertising, which is finding your own platforms and releasing your own research. And when that comes to mind, it’s, it’s a client we worked with, it’s Google, but we found in this study a really fascinating stat that 72 percent of global executives would admit to greenwashing and they admit that other people do it and it’s because they don’t have metrics, like all the metrics are squishy, all the standards, and it actually exposed a problem.

And I thought, this is a really great thing we should put out. And so we put it out and it spurred a significant amount of debate. That’s a topic that was important to Google. And they said, yeah, let’s let it fly.

Emma Varvaloucas: Was there ever a time where what you expected to find as an answer was just like completely the opposite?

John Gerzema: Totally. One of my favorite ones this past year was we do a corporate reputation survey. It’s consumer based. It’s mainstream. So instead of asking elites, we just ask ordinary Americans that buy products and services, what are their favorite companies, which companies do you admire the most? And there’s sort of different dimensions that make the ranking, but it’s really revolves around character trusts and ethics are sort of the big drivers.

And number one was Patagonia and number five out of the top 100 was Chick fil A.

Emma Varvaloucas: No, really?

John Gerzema: And this is all Americans, right?

Emma Varvaloucas: Really?

John Gerzema: And so I’m trying, first of all, to imagine these two companies like at the same corporate picnic. And I’m just fascinated why all Americans would like rate those two companies because their ideologies are so completely different.

And what comes out in the data that’s actually really, I think, encouraging is that people didn’t agree with their platforms, but they respected them for upholding their values, for sort of walking their talk. And I kind of looked at it as an opportunity for a conversation between these two bipolar sides, because we have another Harris data right now, all the toxic politics that have kind of tinged AB InBev or Disney or Target, other companies that got caught up in a lot of this political toxicity. These two companies actually are liked by both Republicans and Democrats because they’re just true to their values.

And they don’t flip flop and they don’t change positions. They don’t virtue serve. They just are who they are. It’s an interesting kind of point because there seems to be this time right now where trust is so low and expectations are so low among the American public for a candidate or even a company that when you are just old fashioned, You can be relied on even if we don’t agree with you.

I thought that was really kind of hopeful.

Emma Varvaloucas: So it definitely wasn’t that so many Republicans were pro Chick fil A. Like, it was that sort of equal amounts of both Democrats and Republicans. Because when I was in university, it was social suicide to eat at Chick fil A. There was one Chick fil A at the food court and you did not wait in line for Chick fil A if you wanted to have friends.

John Gerzema: Yeah, I’m looking at the data right now. So Chick fil A was number five among GOP. The GOP also had Patagonia, number three.

Emma Varvaloucas: Okay.

John Gerzema: Which is really interesting and it cuts across rural and urban. Democrats are less forgiving of Chick fil A. They’re more, they have them, much further down the rankings, but it’s just fascinating that, I’m just going to hold this up for you guys.

I know this is bad, bad form.

Emma Varvaloucas: We’ll try to explain this for audio.

John Gerzema: So this is Chick fil A and Patagonia and how they just effortlessly surf across sort of different political parties, different ideologies. So they are kind of highlighted. And you can just see they’re top 10 with everybody.

Zachary Karabell: Part of that is the Democrat issue with Chick fil A is the corporate supportive conservative causes, religious causes. In terms of employee satisfaction, Chick fil A is off the charts, certainly off the charts compared to mass employers of minimum wage or entry level workers compared to Walmart or the rest. So the Walmart’s improved mightily i think over the past ten years. Walmart’s come up, I don’t know where they show up in your top one hundred but they would have been way way way down twenty years ago and i think they’re probably upper quadrant now is that right? Do you know offhand where Walmart is?

John Gerzema: That’s right. Yeah they’ve improved significantly in the last ten years.

Zachary Karabell: Simply by paying attention to their workers. I mean Walmart cynically at one point realized that if you don’t pay your workers enough money to shop at the stores that they’re working at, particularly for Walmart that has like a million and a half employees, you’ve actually, you’re losing a customer base. You might as well pay them enough to at least buy your own stuff. But Chick fil A, yeah, shit is very, very high. So is Patagonia on obviously different metrics than the college lunchroom, right, that you talk about, Emma, because no one’s thinking that about, are they treating their employees well, but it’s hundreds of thousands, I don’t know what the number is, Emma talked about and asked about the people are pissed off and feeling negative about all sorts of things, direction of the country, economic security, immigration, kind of down the list. There’s um, nothing other than the 69 percent who feel that things are sufficiently bad, but therefore, they are likely to get better, which is a form of optimism, but it is an optimism informed by pessimism. Is there anything in the polling over the past 20 years that would give you any sense as to, why of this progression?

I mean, I don’t know. Maybe in 2019. Do you recall people probably were feeling okay about the economy? It’s been a long time before people in any significant numbers felt good about the direction of the country, right? I mean, that’s it’s been way better than it is now, but it hasn’t been good for the past 15, 20 years.

John Gerzema: No, that’s right. And those, Zachary, have moved lockstep with our growing political divide. What basically happens on these questions now is that they’re highly geared toward being anti incumbent. So, in this month’s Harvard Harris poll, 62 percent of Americans say that the country is on the wrong track, and that includes 32 percent then saying it’s on the right track.

But, if I had shown you that data four years back. It would be nearly identical, but it would be flipped where Democrats would say it’s on the wrong track. So it’s high Republican numbers that are creating that. You only have 18 percent of Republicans saying the country’s on the right track and 55 percent of Democrats.

So that’s the biggest issue. I think it’s less accurate rather to ask the big macro questions because you’re going to get this political response. So give an example right now. 60 percent of Americans say the economy is on the wrong track versus 34 percent rather say it’s on the right track. But when you ask people how do they feel about their own personal financial situation, right?

So we’re taking that out of that voter mentality. It’s at 28 percent improving and 27 percent just as well off. So you put those two together. Now suddenly you’ve got more than 50 percent of the country, 55% that now say the economy’s looking better. Zachary, to your point earlier, you know, about how you frame a question and how you ask it, the more personal you can be in these questions and the less kind of vague and macro and political, the better answers you’re going to get.

We do see a lot of hope on the economy and who knows, it’s way too early right now to be projectable in terms of the election. That’s another really important thing. I just… major caveat, polling is only as good as right now. Polling doesn’t tell you the future and doesn’t often do a good job of even telling you the present sometimes.

But in this instance, what we’re seeing is some really nice upticks in people’s confidence on the economy. We just released a survey with Axios, who we call Axios Heritable Vibes. And two thirds of Americans think 2024 are going to be better than 2023.

Emma Varvaloucas: Yeah, it always, there’s always a bit of a lag, right?

Between the actual economic numbers and how people feel about them. That reminded me to ask you previously, why you said there might be signs that there is a secret Biden voter out there in the way that there used to be a secret Trump voter.

John Gerzema: Well, there’s a couple of things that are a little concerning in the polling as we kind of think about moving forward to the election and a couple of things that we’re seeing happening.

One is demographic shifts and voting patterns among black, Latino, and Asian communities towards Republican candidates. That needs to make sure that the pollsters sort of go that way. But when you get into those biases, this sort of shy Biden voter is just what happens when there’s all this speculation of all this criticism of Biden being too old and too out of touch.

There’s so many other issues that still might bring people to the polling booths, right? It’s obviously abortion and Roe v. Wade. It’s, you know, immigration. It’s the economy. So one of the places we’re really interested in is like, are these left leaning, younger demographics that are being highly critical of Biden right now on Gaza and Israel?

Are they going to sort of do what Trump did, which is like galvanize these voters and get them there, and the pollsters aren’t going to pick it up.

Zachary Karabell: So I want to pivot back to something you said before about your bot problem, or making sure that the person who’s answering your online question is a person. I’ve tried to avoid asking questions about AI just because everybody seems to be incessantly asking questions about AI in areas where it really doesn’t apply, and it’s completely speculative.

But certainly in the polling world of trying to gain things from essentially online presences, that will be an issue or certainly probably already is an issue for you. What do you see going forward about your ability to authenticate a human in your responses?

John Gerzema: Yeah, no, that’s going to be a significant issue.

And we’re working right now to take our fraud protection up another level and match AI with AI to make sure that we’ve got a personal identification, whether it’s a web address or other firm to make sure that we’re not getting botted. And this is going to be a massive, massive issue with identification.

I think the thing that we’re doing on the positive side, we’ve just launched a product we call Quest DIY, which is sort of do it yourself sort of polling where you can kind of create your own survey. It’s like a, like a survey monkey, but it’s AI driven. And the cool thing about it is, is it allows you to sort of get your polling questions going and you get improvements.

So you get suggestions from AI to sort of help make them stronger. And look, the way I think about all this stuff, it’s kindling wood. It’s just a great way. It’s not the answer, it’s not the fact, it’s not the creative insight, it’s not the big idea, I don’t think, not yet anyway, but it’s sure a great start, and so if you’ve got thoughts and ideas, if you’re trying to write a poll, this is an incredibly great way for us to use it, and I don’t know, we’re seeing a lot of progress on it in our business.

Zachary Karabell: Yeah, maybe elaborate on that for a moment, because I know there was an attempt a bunch of years ago, especially a bunch of Argentine economists, and after the collecting government data about inflation and other things became so politicized, it was a bunch of MIT people, they found a way to just scrape data from prices that were posted online as a proxy way of, You can’t collect official inflation statistics, there’s enough price data out there that if you could find a program that would scrape the data, you’d essentially come up with a very accurate read.

Is there a polling equivalent, meaning if you can’t, or maybe juxtaposed to it, where you have a question and there’s so much between whether Reddit and message boards and all the stuff out there that you could essentially use algorithms and AI tools to collate existent online data to get at some of the same answers?

John Gerzema: Absolutely Zachary and we are right now working on a project to take our 60 years of Harris Poll archives of our surveys and try to do exactly that. Because we’ve got just wonderful treasure trove of data on major social issues in America, right? Gun control, racism and equality, freedom of choice versus, you know, pro-life, pro-choice.

And we’re just interested to try to go back and use that to try to understand how the social attitudes have changed. And right now it’s all sitting on like PDFs of old surveys that were done on paper in the 60s. I’m really excited by that because I think it could really provide incredible, and we’re not the only ones doing it, obviously, but just, just incredible insights that might help us make stronger, smarter polling questions in the future.

Emma Varvaloucas: Could you scrape, how Zachary’s asking, from external data? So not data that you’ve already polled for internally, but like he’s saying from Reddit or TikTok comments or something like that, or is it, it’s just completely not hygienic in a way, as far as data goes.

John Gerzema: We’re not doing that per se. We have companies like in our larger group that are looking at those projects.

I mean, I think the big important thing is how do you train your language learning models on the right data so that you don’t obviously have garbage in, garbage out, and you don’t have bias and you don’t have something that comes in that really, as you say, sort of taints your, your data stream. And like, we’re going to have to just for that, even go to this Harris thing, because of the way that they asked polling questions in the sixties, I mean, some of them guys are, are like cringe worthy, right?

So that’s always going to be a problem, but you know, like you said, with your example, with. Argentinian inflation. That’s just a really interesting way to have sort of, you know, behavioral data versus just stated data.

Zachary Karabell: Right. Have you been able to discern, you know, one thing I’ve noticed Is particularly pronounced really over the past decade, probably it’s been true for longer, is the disjuncture, you mentioned this about immigration, that a lot of people don’t have direct experience of whatever the immigration problem is, but because it’s constantly out there as a problem, they’ve internalized it as an act of concern.

I’m not saying it isn’t, I’m just saying it’s a difference between personal experience and public perception or public awareness. There’s always been a disjuncture between let’s say how people feel about Congress and how they feel about their congressperson or how they feel about schools and how they feel about the particular school their children are at or how they feel about the economy versus how they feel about their particular business.

Has that split gotten wider? Meaning there’s just a lot of people will say that their families are fine. Their kid’s school is okay. Their job is decent. Their employment level, they feel okay about X, Y, and Z, but they dislike or distrust the general version of the same specific.

John Gerzema: Yes, I mean, everything generally is trusted more at the local level, and as you expand out, it’s mistrusted when it goes macro.

That’s for the news media, that’s for politicians, that’s for government agencies. I mean, it’s really fascinating that local, like, if we’re going to get anything done in this country, like you’re going to have to do a grassroots up because that’s where the trust is. I think that that’s like really interesting.

And particularly like when we get to the macro news, we just did a poll on misinformation and we found that 62 percent of Americans say they’re cutting back on their news consumption in order to protect their mental health. And so it’s just that the national news media has just become so sort of tainted with perceptions of bias that it’s not only an echo chamber, it’s now a public health crisis.

And that’s the same with government. So I don’t know if that answers your question, but you know, definitely that’s what we’re seeing in our patterns.

Zachary Karabell: Emma did a piece early on in The Progress Network about how to read the news without losing your mind, kind of apropos of the, you know, how do you actually digest the news without going insane or driving yourself crazy as the case may be, so kind of very much in this line of, Most people’s response to that is to simply tune out, right?

Because there’s no, yeah, they don’t, because I haven’t read Emma’s guides to the guides to the dispossessed of how to digest it and stay okay with it. It’s sort of an all or nothing approach. It’s either over indulgence or abstinence.

John Gerzema: I got to read to your post Emma. Sounds great.

Emma Varvaloucas: I’ll send it along. I’m working on a book on it too.

So.

John Gerzema: Oh, nice. Awesome.

Emma Varvaloucas: I think it’s the perceptions of bias is a huge thing and I think it’s also just a feeling of helplessness because we’re just inundated with so much news that we have absolutely no control over from places that 50 years ago we wouldn’t even have seen in our lives, you know, so I think that does a decent amount of making people feel, as you’re saying, maybe great about my personal life, about the world, I feel constantly terrible.

Zachary Karabell: I’m really interested in the rise and fall of company reputations. 15 years ago, Google, Facebook. I don’t know about Amazon, but so, you know, Google, Facebook, Apple were seen as good companies by people, right? There was a high degree of trust since, I remember this at the time of the financial crisis, right?

In 2008, 2009, if you’d asked people which companies they were most negative about, they would have listed Goldman Sachs, Bank of America, a series of financial institutions. And now today, I don’t know if you’re finding this, but I think the things that most people are agitated by are big tech companies of one form or another.

We’re recording this, by the way, during the week when the House of Representatives passed its first effective step toward banning TikTok, which may be an anti Chinese move, but it also has a degree of just generalized social media. These companies are bad companies doing bad things. Do you have a sense of why that is?

Or is that too just another one of the press? The PR got worse and then people’s attitudes followed? Or do people’s attitudes get worse and then the PR followed?

John Gerzema: I think it’s a little bit more of the latter, but we have seen over the last five to seven years in our reputation survey, it’s the Axios Harris Poll 100 that tech is sort of bifurcated into two camps, sort of good tech and bad tech.

And so, I’m generally speaking, but the good tech are tech companies that make things, so, you know, Samsung, Consumer Electronics, Microsoft, but if you’re a social media company, you’re at the bottom of our survey. If you’re TikTok, if you’re Meta, you know, if you’re, I guess Instagram rolls up under Meta, but those companies are sort of seen as harmful to society and not being productive, as it were. The other really interesting thing that’s happened again, my low tech way, I’ll show you, but we tracked 15 years of Harris data looking at companies that had had a huge crisis. So this has got BP, Wells Fargo, Volkswagen, Chipotle with E. coli. And generally what ended up happening is they all basically had a U shaped recovery, right?

They had a crisis and their reputations came back. So they’re all kind of back to the same place they were. But the new thing we’re seeing is an L shaped crisis, which here we have Disney, AB InBev, and notably Meta, and Meta has the longest timeline, but what you might be able to see in my wonky chart here, is that Meta hasn’t recovered, and the reason these companies are all facing these L shaped crisis where their reputations haven’t come back, is they’ve fragmented between Democrats and Republicans, so they’ve entered a new territory, a political territory, where, for one reason or another, you know, with Disney, it was the Don’t Say Gay Bill, with Meta, it was being perceived as sort of anti Republican in Cambridge Analytica, ABI was obviously last year with Bud Light, you just end up in a potentially a longer term crisis because you just alienated half your audience by sort of challenging them on a political level, which nobody wants to be challenged on when you just want to have a beer.

Emma Varvaloucas: Interesting.

Zachary Karabell: Well, John, I think we are at our time, but I, I really want to thank you for the conversation kind of wide ranging from tech to consumer to politics. And I urge everyone who’s listening that the weekly, what’s the exact name of the weekly service so people could look it up?

John Gerzema: America This Week.

It’s under my name, John Gerzma from the Harris Poll at LinkedIn.

Zachary Karabell: I would definitely check that out. It’s a really, really good snapshot of just attitudes writ large. And obviously looking at it, not just week to week, but over time is particularly helpful just to see some sense of trends, but really appreciate the time.

Definitely appreciate the work and the data and we’ll see how this year plays out and public attitudes and both tech regulation on political outcomes and whether or not sentiment matches reality.

John Gerzema: Really enjoyed it guys.

Zachary Karabell: Thank you.

Emma Varvaloucas: Thanks John. Well, that was a fun one for us data nerds here at The Progress Network.

It was an interesting little tidbit at the end there that polarization has come for company crises right? Like I hadn’t even forgotten that Wells Fargo had a crisis, completely forgotten about BP until he mentioned it. But I guess nowadays, once you go down, you stay down.

Zachary Karabell: Although the fact that you’d forgotten about BP, whoever is, you know, listening, if BP’s crisis PR team stumbles upon this episode, they’ll be like…

Emma Varvaloucas: My friend is an advertiser for them, that’s probably why.

So like my, my most recent reference point doesn’t have to do with the oil slicked pelicans and so on and so forth. Good for BP, not good for us.

Zachary Karabell: This constant disjuncture between personal attitudes, local attitudes, personal, like your, your sense of how your own life is doing versus the sense of the collective and the degree to which the skew is personal positive, collective negative.

There’s a generalizable statement that tends to be the case. And that’s an even more stirring indication of most of people’s experience of a collective reality is filtered through whatever we call the news and or social media, right? Because we don’t have a direct experience of it. We only have the experience of what we hear online or see online or read.

And so our sense of reality beyond our media is shaped intimately by a few types of medium, but not our own senses and not our own life experience. And if that’s largely being told in the negative, it’s likely that we’re gonna have a negative perception, right?

Emma Varvaloucas: Yeah, I was thinking about that as we were discussing that, that the ideal situation would be that we feel positive about our personal situations and we feel positive about the collective, right?

Zachary Karabell: Right.

Emma Varvaloucas: So we’re not in the ideal, but I’d rather have the situation that we have now than a situation in which you feel really negative about your personal situation because that likely means that there are real reasons for that and feel positive about the collective, which would just be like a delusion that would be just as harmful, right, than the one that we have now, but actually I think I’m saying, I think that would be more harmful than the one that we have now. I’m kind of pulling for the, at least in our personal lives, we’re doing kind of swell and that might affect the intensity of how negatively we think of the collective.

Zachary Karabell: And the one exception, which we didn’t talk about with him is of course, people do, at least in the United States, feel economically insecure by large measures.

And that’s a very personal one. Meaning, do you personally feel economically insecure? Do you have enough money to meet a crisis of one from another? And the answer is largely no. So those are personal ones about one’s experience of what’s going on economically, and I do think that those have a high degree of legitimacy.

What’s interesting is it’s not entirely clear, or rather it’s not clear at all that one’s ability to do that 30 or 40 or 50 years ago was any better. What clearly has changed is that your expectation that it not be as bad has risen, meaning a tenable level of insecurity in 1970 is not a tenable level of insecurity in 2024, if you are your, your average citizen.

Emma Varvaloucas: Yeah, and what’s really interesting to me about this is that people now are starting to bring up the economic insecurity of Gen Z, but it’s actually a repeat of the pattern that we saw with millennials, I think. I don’t think things have materially deteriorated between Gen Z and millennials that much, and there was a very long time that millennials were saying that we’re economically insecure and there were reasons for that and that wasn’t illegitimate, but now it shows that millennials have caught up wealth wise with other generations where they were at this age, and in fact, surpassed them. So there actually is some economic data that says that in fact, we are in a better position than we were previously. And I think that some of the insecurity, particularly around the younger generations, just has to do with the fact that you’re young, you’re not making that much money yet.

A lot of people have college loans, but that doesn’t mean that’s going to stay that way forever.

Zachary Karabell: Yeah. I mean, it’s hard for me to believe this now, but the onset of the 2008 2009 financial crisis is, you know, 16 years ago or 15 and a half years ago. And if you were a millennial who’s kind of first brush with the external world was that crisis, you know, that certainly is going to inform your attitudes for a whole bunch of years because it was a bad couple of years and it took a long time for things to quote unquote normalize.

The other interesting thing about the John and Harris poll is, at least in the private realm, right, they’ve kind of gotten around the whole issue of how do you reach people by simply paying for panels, which remains a problem with political polling, right, because one of the hardest things now for a lot of the pure political pollsters is actually finding people to answer the questions, and not just online, like I guess YouGov does, Tens of thousands of online surveys, but you never know how representative those tens of thousands who are answering those are.

Emma Varvaloucas: I’m still a little bit unclear. Like, I know that if you offer people, I’ve read that opt in online surveys are notoriously unreliable because they attract trolls, because who are the people that are hanging around online? It’s the trolls, right? And that’s why you get really crazy numbers on some polls, like 30 percent of Americans are Holocaust deniers, but like, in fact, they’re not.

So I guess I still don’t understand like what the correct way to attract someone online is that you like, you buy a list off someone and you email them and you offer them a hundred bucks. Is that how they do it? I’m not sure.

Zachary Karabell: And look, this is going to be the year of obsessive following of polls politically, but frankly, all that’s going to matter is the polls in like six states for the US presidential election. National polling is going to make very little difference. It doesn’t really matter if Trump is 4 percent ahead of Biden or Biden’s 6 percent ahead of Trump. It matters how either of them are doing head to head in Georgia, Arizona, Wisconsin, Michigan, Pennsylvania, and Nevada. Anyway, gonna be a fun year.

Everyone should check out the Harris Poll information. They’re one of the more… at least I find one of the more insightful, nuanced polling organizations, partly because they’re not driven primarily by a news electoral cycle, where they’re supposed to give some sort of binary horse race. They’re much more informative polls about attitudes than they are horse race polls, and I think, therefore, more interesting to read.

Shall we turn to our rapid fire? Things you should have been paying attention to?

Emma Varvaloucas: Rapid fire positivity. Yeah, let’s do it.

All right, so let’s take a look at some news that probably went under most people’s radar, starting with the Caribbean Island Nation of Dominica. When’s the last time you heard a news item about the Caribbean Island Nation of Dominica?

They have overturned a ban on consensual same sex activity. Good for them. That has been on the books since British rule in the 1800s. And what’s neat about this is that it’s part of a bigger trend with Caribbean nations in the past few years. So Barbados, St. Kitts and Nevis, Trinidad and Tobago, and a few other countries have also overturned this ban.

It’s not completely legal all over the Caribbean, but it’s kind of creeping and trending that way. So it’s nice to see.

Zachary Karabell: Yeah, there’s a lot actually truly going on between Caribbean nations around sustainability and environmental change. You know, island nations tend to have a strong self interest in doing something about climate change because of obviously rising sea levels and dependency on tourism and the degree to which those two things could not easily coexist.

So these are countries that have not usually received much particular attention in global affairs, but have become, we once talked about American states as being incubators and experiments of democracy. A lot of these island nations have become oddly, oddly in the sense of their history, not oddly per se, creative in the way in which they’re dealing with certain problems and progressive, small, small P progressive in how they come up with solutions.

Emma Varvaloucas: 100%. Speaking of that, on the sustainability line, Ember just released their global electricity review and the world hit a pretty big renewable energy marker last year. 30 percent of the world’s electricity was produced by renewables last year, which is honestly more than I would have expected. A third.

Not too shabby.

Zachary Karabell: That is a lot.

Emma Varvaloucas: They think that fossil fuel generation will fall in 2024 and then like really start to decline post 2024, but I’ve been waiting around since about 2022 to hit peak fossil fuels. It’s been kind of plateauing and people are wondering when is that peak going to be and I’m hoping it’s going to be right now.

Zachary Karabell: Yeah, I don’t think we’re getting to peak oil consumption anytime soon. What’s declining radically globally is coal, oil, and natural gas. I think we’re not going to plateau. I think a lot of that is because coal is the one that’s shrinking precipitously.

Emma Varvaloucas: Coal is shrinking precipitously.

Zachary Karabell: Which is good because coal is particularly dirty, natural gas is far less so.

Emma Varvaloucas: So, moving on, last but not least, uh, this is a fun one. So we definitely talked about this on the podcast, the Herculaneum Scrolls that were decoded in this challenge called the Vesuvius Challenge, and the winning team used the Herculaneum AI to figure out what was written on these scrolls that they haven’t been able to unroll for, I think, centuries now.

I think they were discovered in the 1800s, because they’re essentially like calcified by lava. Now that they figured out a technology to read what’s on the inside of the scrolls without physically unrolling them, they’re discovering all sorts of fun things. One of them is the exact location of Plato’s burial place.

They didn’t know exactly where he was buried. And they also found this fun fact that apparently there was a slave woman that was playing the flute for him on his deathbed. And he said, while running a high fever and actively dying, that she had a scant sense of rhythm. So that was one of Plato’s last words.

Zachary Karabell: Wow.

Emma Varvaloucas: Yeah.

Zachary Karabell: Kind of like, like, bitchy at the end, you know? Like, here he is, all he can do. And that’s just amazing. But whatever. It’s very cool. I mean, it’s definitely very cool what we can discover. And I wonder if this will lead to a whole slew of both new data, we have like ruins and rocks and some writing, but it’s a lot of piecing together little fragments.

And, you know, the idea that this can unlock more texts and that that can enrich our sense of what happened thousands of years ago is a really cool idea.

Emma Varvaloucas: Yeah, I think there’s going to be a lot of like, we thought this, but really this, because there’s, it was one of the biggest libraries at that time.

So there’s a enormous amount of information to be discovered.

Zachary Karabell: Stay tuned.

Emma Varvaloucas: Yeah. So that’s it for today. Hopefully one of those things got you into a better mood from all the doomscrolling than we normally do.

Zachary Karabell: So please send us your comments. Send us your tired, your hungry. No, we are not. We are not Emma Lazarus and or the Statue of Liberty, but we are What Could Go Right.

And much like Emma Lazarus and the Statue of Liberty, we want to welcome hopes and dreams and possibilities. So if you have those and you want those to be examined more, send us your ideas. If you think that we are glossing over things that we should focus on more intently, send us those as well. And thank you for listening.

We’ll be back with you next week. Thank you, Emma.

Emma Varvaloucas: Thanks everyone. Thanks, Zachary.

Zachary Karabell: What Could Go Right? is produced by The Podglomerate, executive produced by Jeff Umbro, marketing by The Podglomerate. To find out more about What Could Go Right? The Progress Network, or to subscribe to the What Could Go Right?

Newsletter, visit theprogressnetwork. org. Thanks for listening.