Polls you might have missed this week
There's more to life than Republican primary horserace polls.
Hello! This is the first edition of a regular thing I’m hoping to do: A newsletter sending you all links to interesting polling stuff you might have missed if your inbox isn’t a nerdy gathering place for all things public opinion like mine. I might as well do something useful with all the tabs I have open, right? What you won’t find here is horserace election polls, unless there is something specific about that particular poll that I’m pointing out.
Link #1: “Politics Isn’t Life,” The Liberal Patriot
My favorite this week is a bit of a hobby horse of mine - making the point that normal people (i.e., non-politicos, life outside the Beltway exists, y’all!) don’t actually pay that much attention to politics. In this piece, John Halpin discusses results from TLP/YouGov’s survey of 3,000 registered voters nationally that show not all that many people actually “do” political stuff.
Here’s the key chart, copied from the post:
Halpin has a good discussion you should read. My one addition would be to say that we need to keep in mind that people are more likely to have over-reported than to have under-reported their participation in these things. Social desirability would lead some respondents to say they’ve done things they haven’t. Also, posting on social media is a pretty low and inactive bar that I’m not sure I would count as real action.
For a more academic read…
This article in the Journal of Personality and Social Psychology on “The Political Is Personal: The Costs of Daily Politics.”
Link #2: Tusk, “Is it still the economy, stupid?”
This is technically not a single survey, but Seth Masket uses a whole bunch of surveys on presidential approval and views of the economy over the last couple decades to think through how the relationship between economy, presidential approval, and incumbency affect re-election chances. Basically, these relationships are probably changing compared to conventional wisdom. Presidential approval exists in a very different space than it used to, per this chart from Gallup:
Also relevant:
Gallup released a historical look at presidential approval that isn’t great for Biden, unless you consider that approval doesn’t move as much as it used to. The best part of this (to me) is the poll count column - during the Obama administration, Gallup was doing daily tracking and had 88 polls in Q2. Gallup’s daily tracker was by phone, which is hideously expensive, but most that exist now are online. I tend to think daily trackers aren’t super useful these days (see link #1 point) and probably contribute to over-surveying the population that is willing to take surveys online.
Link #3: Marquette Law School poll on SCOTUS
There’s a ton of really great stuff in this poll release, but the part I want to direct your attention to is how much people have heard about the big Supreme Court decisions from this past term.
Just over half of Americans have heard a lot about the two big cases from the end of the term — student loans and affirmative action. Attention drops precipitously from there, which has massive implications for opinion on the cases:
Look at those “not heard enough” numbers, even for arguably the biggest cases of the term. Now go back to link #1. Anyone who wants to credibly gauge opinion on these topics needs to take knowledge into account. If there’s not a way for people do say they don’t know, ignore those numbers.
Link #4: BrightLine Watch on Expert vs. Public Opinion
This survey deals with a lot of questions about small-d democratic norms. The headline, and most of what you probably saw from the survey (if anything) were the numbers about Trump’s indictments. But what I want you to notice is the chart at the end comparing “expert” opinion (a sample of academic political scientists) to public opinion. For these issues I’m willing to say that the experts probably know much more about where the country actually is, so it’s fascinating to see where the public thinks things are in comparison.
Link #5: The Liberal Patriot & AEI on “Double Haters”
This post from Ruy Teixeira is an example of a unique way to look at horserace data that I find worth additional attention. From a survey of 6,000 Americans from the Survey Center on American Life, he looks at the people who dislike both Biden and Trump, and shows that in a Biden vs. Trump matchup, they swing more toward Biden. But if it’s Biden vs. DeSantis, they swing more toward DeSantis. All of this is very fluid right now, and I would not be surprised if these numbers change tremendously over the next 15 months… but Teixeira (and colleagues Nate Moore and Karlyn Bowman, in another post for the American Enterprise Institute) are absolutely correct that some of these people will vote, and those votes will matter in swing states. Here’s the key chart from the AEI post:
Link #6: Marist Poll on Trump
Just a quick one to wrap up the post: Asking “from what you have read or heard about the investigations into former President Trump, do you think he has done something illegal, has done something unethical but not illegal, or has done nothing wrong?” Marist finds that Republicans who say Trump has done nothing wrong dropped 9 points (50% to 41%) since our June poll. Of course, the poll was conducted before the newest revelations: “Note the survey was fielded July 24th through July 27th, prior to the announcement Thursday afternoon of additional charges against Trump in the documents case.” The pollster’s constant frustration: news cycles don’t stop for us to poll and get numbers out!
I really enjoyed this article, Natalie. It sent me to useful articles and perspectives on public opinion, the vast majority of which I had not read before today.
So I comment to encourage this type of article in the future.
Thanks for the shout out. Great collection!