Loading summary
Transcript31 lines
- [00:00]
A
Quick question. On a scale from one to I'm about to be turned into a paperclip, how worried or freaked out are you about AI after a week of think pieces?
- [00:14]
B
Honestly, right now I'm not. And I'll tell you why. I have been working with Jamil to like build some AI tools to help like find polls in the, in the news and process the data. There is a team of people literally trying to make AI do my job and it sucks.
- [00:36]
A
All right, well, there you go. You can't be automated, Mary. You're too good.
- [00:41]
C
Mary and I don't disagree often, I think on this podcast, but on this particular question, I think we disagree pretty strongly. My level of worry has not changed in the last week since like the new models have been released, but I am generally pretty. I'm. Worried is the wrong word, but like, I think we are about to go through something quite weird. And I'm concerned on a personal level because of what I do for a living and I'm concerned on a political level and I'm concerned on a societal level sort of. There are many different levels of concern here that I, that I have.
- [01:11]
A
I mean, call him Susan Collins. The boy is concerned. Hello and welcome to the GD Politics Podcast. I'm Galen Droock and we've got a lot to talk about today. I believe we have our first ever emergency edition of Good data, Bad data or not data. Gallup announced this week that after 88 years in the field, it will stop tracking Americans approval or disapproval of presidents. Their final approval rating for President Trump was just 36%. Gallup is no longer asking Americans how they feel about the president, but plenty of pollsters still are. And that will be useful for two topics that we're going to discuss today. The showdown over Department of Homeland Security funding and the political fallout over the Epstein files. We've also got election news to check in on. The Democratic primary In New Jersey's 11th district has served as a microcosm of Democratic Party drama. A little known progressive organizer won the primary after an APAC backed group spent 2.2Million dollars attacking a moderate pro Israel former congressman. Yes, you heard that correctly. Susan Collins also formalized her bid for a sixth term in the Senate this week, which means another opportunity for us to talk about the race for control of the Senate. We will absolutely take it. And dear friend of the pod, Mary Radcliffe, did a deep dive into whether support for Trump is crashing amongst young men. So a lot to cover. If you're wondering why the current freakout over AI is not on that list. It's because I'd already planned an AI episode for President's Day, even this week's events. So just hang tight on that one. We will get to it. But here with me to discuss all we have for today is dear friend of the podcast and head of research at 50 plus one, Mary Radcliffe. Welcome to the podcast, Mary.
- [03:01]
B
Hey, Kaylin, great to be here.
- [03:03]
A
Great to have you. Also here with us is senior data scientist at the Washington Post, Lenny Brauner. Welcome, Lenny.
- [03:08]
C
It's great to be here. Thanks for having us.
- [03:11]
A
Always a pleasure. So let's, let's dive into this first ever emergency Good data, bad data or not data? Gallup said on Wednesday that it will no longer track presidential approval and favorability ratings for actually all political figures. They described it as, quote, an evolution in how Gallup focuses its public research and thought leadership. They said the context around these leadership rankings has changed and that they are now widely produced, aggregated and interpreted and no longer represent an area where Gallup can make its most distinctive contribution. They also said our commitment is to long term, methodologically sound research on issues and conditions that shape people's lives and that that work will continue through the Gallup Poll Social series, the Gallup Quarterly Business Review, the World poll, and our portfolio of US and global research. When they were asked if this was a response to any political pressure, they said that it was not. So, Mary, is this good data, bad data or not data?
- [04:17]
B
Well, I, I mean, Gallup in general, I would say is good data. They do really interesting work. Losing this series is obviously, sadly, a not data situation. We're not going to get more data from Gallup going forward. And you know, while they are right that we do have lots and lots of pollsters out there measuring public opinion about political figures, one thing that I think is pretty disappointing here is that this is a very long running series, right? So if you are doing analysis, it's useful to say, okay, we have the same methodology over time that we can compare X, X and Y. That's gone. Right. So can I compare like let's say a YouGov polling series on Donald Trump's approval to Gallup's polls of, I don't know, Truman. Truman, Sure I can. But those are kind of different, right, because they're conducted in different ways in different contexts and things like that. So the long running is vanishing is pretty disappointing. I will say, though, I'm sort of unsurprised that Gallup made this move in some ways a few years ago. They really changed their focus of their operations. They do a lot more global work than they used to. They're developing like methodologies for different parts of the world, different communication environments all over the world. So they've been expanding their focus on global research and slowly dropping the American political research. So they stopped pulling presidential elections a while back. They used to have a daily Gallup job approval number for President Trump in his first term that they dropped that as well. So like, slowly they've been reducing the amount of focus on American political surveys.
- [06:09]
A
And is that, can you credibly say that, that just because a lot of people are doing that work. Work. And so it's not a differentiator?
- [06:17]
B
I mean, absolutely. Especially the American polling industry. Right. I mean there's in the 50 plus one database, there's like 2,000 pollsters or something. Right? So yes, we, we've got lots of people doing that work and lots of people doing very credible work in that arena. So I, I actually do find it believable that they're like, this is not a value add for us. And it's also the kind of area where they're pro. They may be taking a lot of incoming if they're printing bad numbers for someone's preferred politician. You know, I'm not thinking about incoming from necessarily like people in power, but just like we get emails.
- [06:58]
A
Okay, Lenny, is this good data, bad data or not data?
- [07:03]
C
I agree with Mary's characterization. It is not data. It is bad for data. And Gallup General produces good data. I feel like that's a nice. It hits all three. I will say listeners of this podcast will know that I generally often rail against polling questions that are entirely unverifiable. I think this is one that is kind of unverifiable. So in some sense, maybe that is part of the thinking here. I mean, I guess. But they got rid of, like you said, Mary, they got rid of their presidential head to head poll recently also, which I guess is the only one that is the most. It's the most verifiable of the bunch. So maybe that is not their thinking. I will say there is a sort of elephant in the room here also. I mean, I realize that this formerly used to be the538 podcast. I realize that who you work for, Mary, and who a lot of us sort of work for. But there is the aspect that polling aggregation has kind of taken the umph out of any one particular poll. The polling aggregators suck the oxygen out of the room of the discourse when it comes to individual polls. And so for institutions, I think that's another reason why they might decide that it is not worth for them to produce polls anymore, because they're not getting the publicity they once got for releasing these polls. Because no one, beyond sort of a small subset of maybe election Twitter, talks about these individual polls. Everyone only goes like, you know, what is the 538 average? Or something like that.
- [08:19]
B
Yeah, that's true. I mean, you do sometimes get headlines around an individual poll, but generally only if it's an outlier, which is really, really bad.
- [08:28]
A
Yeah, we've all. We've all learned that lesson a couple times over at this point. But you make an interesting point here, which is that in many cases, these are research firms that are trying to turn a profit. And so, you know, if it's at a university, it may be not for profit. It may be motivated by, you know, academic research rigor, et cetera. But for research firms, they want to get publicity so that they can drive private clients to their other products and do a bunch of research. And if they're not getting that publicity and are only sort of, I don't know, maybe turning off half of the country because they're putting out bad numbers for one side of the political spectrum, that may even hurt business. And so we don't oftentimes talk about the biggest business side of polling here, but it is real. These folks are trying to make money off of learning how Americans or people across the world think about things. And so I. I have to imagine that this is, in part, a business decision.
- [09:24]
B
Yeah, I mean, if you think about the kinds of surveys that Gallup puts out that get more headlines these days, it's things like the Economic Wellbeing Index, for example, that recently came out. And, and there I saw, like, a bunch of articles about the. The numbers in that. In that survey. So that. That makes sense.
- [09:42]
A
Okay. All right. Well, we say goodbye to 88 years of Gallup presidential approval tracking, although we can maybe rationalize it through. It's a business decision. There's lots of people in the space. It is still a loss for the polling community and a loss for good data. So we are sad to. To say goodbye those 88 years of presidential approval. That data set is coming to an end. We shed a tear for it. Let's move on and talk about some of what is happening in Washington. The Department of Homeland Security will shut down this weekend if a funding deal isn't reached by Friday. It's now been almost three weeks since Alex Preddy's killing, and we mentioned at the time that we probably needed to wait a couple weeks to see where public opinion landed. Once the dust had settled a bit, we've had that opportunity. How do Americans feel about ICE and immigration enforcement as Republicans and Democrats debate a deal in Congress? Money.
- [10:40]
C
That's a great question, Galen. So while we have had a bit of time for, like you said, the dust to settle, we just haven't had that many polls that ask about this question in the last two or three weeks. From the polls that we do have, it seems that the changes that happened in public opinion has sort of remained relatively Constant. In the YouGov poll, Trump's handling of immigration was 13 points underwater, which is his lowest net approval in the second term and second lowest over both terms for immigration in particular and abolish, ICE was 47% approved, 42% disapprove or oppose. That is basically no change from the recent spike that we saw in the YouGov poll prior to this one. So that seems to have remained relatively constant. The NBC poll, with a CAVEAT that's already two or three weeks back, ICE approval, you know, 66% somewhat disapprove or strongly disapprove of that, 55% strongly disapprove. So there's a large chunk of people who strongly disapprove, and 34% somewhat approve or strongly approve. There is a bit of a difference. For that NBC poll, when they asked about abolishing ice, because they actually gave three options. They gave abolish, continue its current form, and reformed. And given those three options, actually a plurality of Americans want to reform ICE, 43%, rather than 29% that want to continue in its current form and 29% that want to apologize.
- [12:04]
A
Okay, so I think that's good context. We talked a little bit at the time about how we see in the aftermath of gun violence. For example, we see a spike in, you know, support for gun control measures. Some of that tapers off as the tension fades. At least that polling tells us that we remain relatively in the political moment that we have been in since Alex Preddy's killing. Democrats are trying to use this moment, in a sense, to make some demands and reform immigration enforcement. Democratic leader Hakeem Jeffries and Chuck Schumer put forth basically a list of demands saying, for example, that they wanted targeted enforcement, you know, focus mainly on criminals. They wanted no masks for enforcement agents. They wanted to require id. They wanted to be more specific about locations where enforcement could take place, like churches, schools, hospitals, things like that. They wanted, you know, Standards of force. They also wanted body cams, but sort of with some assurances that they wouldn't be used for tracking people. A list of 10 demands, basically. And I'm curious, you know, the White House got back saying, okay, we're going to do body cams. They had already, before these demands were released, suggested that they would do body cams. You know, are Democrats on solid footing with the public in terms of the demands that they have put forth?
- [13:29]
B
Yeah, I think generally, yes. So I went through those 10 bullet points from Jeffries and House Democrats and I looked for polling data asking about each of those items individually. Specifically, I'm going to Cite data from four different polls here. YouGov economist conducted January 30th through February 2nd, a GBAO poll conducted for some Democratic groups February 6th to 9th, that NBC News poll that Lenny was referencing earlier, and a Navigator research survey from January 29th to February 1st. Now, I want to flag of those four polls, two of them are from Democratic leaning organizations, the GBAO and the Navigator Research Survey. But that's because Democratic organizations are the ones interested in asking about these questions right now. So, like, keep that caveat in mind. But I found polling questions that asked about eight of the 10 items on Jeffrey's list in some way or another. And of those eight, none of them are below 60% support with the American people. And some are much, much higher. So, for example, on that targeted enforcement requiring judicial warrants, GBAO and YouGov economist both show 69% of Americans support that. You mentioned body cams. That's one of the highest levels of support. YouGov economist 84% of Americans saying they support body cams, use of force standards. Navigator Research asked it a couple different ways. They found support at 76% or 70% depending on how they framed the question. So, and I could go through more of these. There's just generally speaking, each individual piece of this plan seems to be quite popular with the American people.
- [15:11]
A
Listeners will also be familiar with the idea that even when, for example, support for gun control measures spikes, they don't actually get implemented. So just because there is public, broad public support for most of these measures, does that mean that we can expect, you know, Democrats to win this fight? I think perhaps we can expect that the government will shut down because Democrats do feel like they have the upper hand here. I said government. What I do mean is the Department of Homeland Security, 96% of discretionary funding has already been passed through Congress. So the 4% remaining for the Department of Homeland Security that is what will be shut down. And, and another caveat here, ICE has already been funded, as we've discussed many times, the one big beautiful Bill Act. So ICE is actually also not going to shut down. But nonetheless, I think it's likely the government will shut down. Polymarket, put it at 82%. Should we expect Democrats to get some, some of these measures implemented through this shutdown process?
- [16:10]
B
I mean, we've talked about this before. It is pretty rare for policy change to come through a shutdown process. And the other thing I would note, you know, like, these individual items are popular, but we see this with stuff with like, for example, the Affordable Care act, where you can ask people, do you support each of these points inside the Affordable Care Act? And they'll be like, yes, yes, like pre existing conditions, yes, whatever, all of those things. And then you ask, do you support the Affordable Care act, which is just all of those things tied together with a little ribbon, and the support goes dramatically down. So while I can say each of these individual items is quite popular, what I don't know is if you ask people, do you support the ICE reforms proposed by House Democrats or do you.
- [16:53]
A
Support shutting the government down over these ICE reforms or something like that?
- [16:58]
B
Yeah, exactly. And that falls far more along party lines. So we've seen polling about the shutdown specifically that really does fall more along party lines. You get much less support for the shutdown than you do for these, like, individual pieces of reform. So my suspicion is that there maybe is a compromise to be reached here. Right. Like, it seems like the Republicans in the administration are willing to compromise on things like body cams. Maybe they can get a couple more of these. But I don't think they're going to get this whole package. I mean, I don't think they're going to get this targeted enforcement requirement for a judicial warrant before they make an arrest. The Trump administration doesn't seem like they would be interested in that.
- [17:36]
C
There's also the point that you raised, Galen, that I think is really important here, which is that the shutdown pain that the Democrats want to enforce here or want to force, you know, the Republicans to feel is actually not unclear whether it really helps the Democrats, because like you said, ICE is funded. The parts of DHS that are going to lose funding are parts that Democrats generally also care about, like the tsa, like cybersecurity aspect, things like that. So it is unclear to me whether Democrats are really able to turn up the heat here in a way that they would like to, even if they feel that they have, politically speaking, the public backing to do so because they feel like, you know, a lot of the things that they are arguing for have broad popular support.
- [18:15]
B
Yeah, that's a great point, Lenny. Like especially tsa, because the public feels it when you shut down tsa, Right. Like, are we going to close airports over this issue? I don't know. The public likes to support things that they like unless it affects them negatively.
- [18:33]
A
Yeah, we'll have to see what happens. There's also the obvious alternative, which is this is a midterm year and Democrats choose to run on this issue and sort of cut their losses at a certain point with the understanding that wholesale reform is unlikely to happen while Trump is in the White House and Republicans control both chambers of Congress. Speaking of things they may or may not choose to run on this fall, let's move to our second Washington based issue. On Wednesday, Attorney General Pam Bondi appeared before Congress and answered, or perhaps you could say filibustered questions about the Jeffrey Epstein investigation. The Department of Justice has now released roughly three and a half million pages videos, images related to the Epstein files that journalists and lawmakers have been combing through over the past several weeks. I'll say here that Trump's handling of the Epstein files is perhaps his worst overall issue. At net negative 34 approval per YouGov inflation is at net negative 28. So you can tell by that number that there are a decent number, like basically. But he's improved.
- [19:41]
C
He's improved. He was at negative 42 in September.
- [19:44]
B
Okay, do you guys want it? I did this really silly thing when I was thinking about this. I at 50 plus 1, I track Donald Trump's approval on every issue that any pollster asks. Do you approve of how Donald Trump is handling blah. I have a database of all of those questions. So I looked at all of the polls since the start of this year, since January 1st and I sorted them by net approval and of the top 10 worst issue polls.
- [20:09]
A
And that's the end of today's preview. Head over to GDPolitics.com to become a paid subscriber and catch the full hour long episode. We covered a lot of ground. How the Epstein files might figure into Democrats midterm strategy, why the Epstein files seem to be having more of an impact in Europe than in the U.S. we also talked about the messy, messy New Jersey 11 Democratic primary, the state of play for control of the Senate and whether support for Trump is crashing amongst young men. We got to all of that. So like I said, head over to GDPolitics.com to become a paid subscriber and catch the whole episode. Paid subscribers get about twice the number of episodes. You can also join in the paid subscriber chat. And most importantly, you keep this podcast going going. When you become a paid subscriber, you can connect your account to wherever you listen to podcasts so you'll never miss an episode. There's a link in the show notes explaining how exactly to do that. Again, head over to GDPolitics.com see you there.