Trump Unpopularity Check
And all the links
I haven’t revisited Donald Trump’s popularity for about five weeks now. Last time I looked at it was right in the middle of a flurry of polls leading up to his 100 day in office mark, and he was mostly doing badly.
He bottomed out soon after that, and seems to have improved some for a bit and the inched back down a little. Let’s put some numbers to it: I wrote on April 26 (publishing the next day) when his net approval via G. Elliott Morris’s excellent new Strength in Numbers estimate had him at -9.8 percentage points; he dropped down to -10.5; recovered to -7.4; and is now at -9.0.
Or at least that’s what the chart says. It’s just as likely that he’s been right around the same point the whole time. Among other things, “net approval” – that is, approval minus disapproval – is a fine number to use, but it seems like more movement than just plain approval, even though it pretty much isn’t. (That’s because relatively few people tell pollsters that they have no opinion, so people are moving back and forth between approve to disapprove, thus doubling the apparent movement.) (Apologies for having some math here for those who don’t like it, and for only a little math for those who want more).
We also shouldn’t take the exact numbers too seriously. The advantage of polling averages is that they help to give us better estimates than single polls. But the fewer the surveys, the more we should expect random deviations from whatever the true balance of approval and disapproval might be in the nation. In this particular case we know that many pollsters wanted readings near that 100 day mark and we’ve had a very slow month of polling since then. The fewer polls, the better the chance that single oddball surveys will make it look as if there was big movement that may not really have existed. On top of that, it looks to me as if some of the lower-quality polls have been a greater share than usual of May polling. So we should have a pretty reliable estimate of where he was in late April, but much less so since then.
Look, all of this may be for myself as much as for anyone else. I’m as prone as anyone to examining the charts at Strength in Numbers (or other such sites), trying to suss out exactly what’s happening. It’s a bad idea when I do it, and a bad idea when you do it.
At any rate...my guess is that Trump steadily lost ground from January 20 through the end of April, and has been relatively flat since then. Either way, he’s still comfortably in second-to-last place among first-term presidents during the polling era, trailing only his own record in 2017. He’s still about 22 percentage points of net approval behind Joe Biden through June 1, but Biden is no longer the worst other than Trump. That place belongs to Bill Clinton in 1993, who is only 12 points better than Trump.
The big story is the same as it was five weeks ago: Trump is very unpopular for a new president, although not anywhere close to as unpopular as some presidents have been later in their terms.
How about some links:
1. Molly Reynolds on the latest twist in Senate procedures.
2. Jennifer Victor at Mischiefs of Faction on the megabill, deficits, and debt.
3. Matt Grossmann talks with John Dearborn and David Hausman about presidential authority.
4. Dan Drezner on undermining neutral expertise at the State Department.
5. Natalie Jackson on public opinion and the courts.
6. Eun A Jo at Good Authority on the upcoming election in South Korea.
7. Chris Baylor on Disney remakes gone wrong.
8. And Dan Nexon on the farce that is “unitary executive” theory.