Gallup Ends Presidential Approval Polling
I question the timing.
The Hill (“Gallup will no longer measure presidential approval after 88 years“):
Gallup will no longer track presidential approval ratings after more than eight decades doing so, the public opinion polling agency confirmed to The Hill on Wednesday.
The company said starting this year it would stop publishing approval and favorability ratings of individual political figures, saying in a statement it “reflects an evolution in how Gallup focuses its public research and thought leadership.”
“Our commitment is to long-term, methodologically sound research on issues and conditions that shape people’s lives,” a spokesperson for the agency said. “That work will continue through the Gallup Poll Social Series, the Gallup Quarterly Business Review, the World Poll, and our portfolio of U.S. and global research.”
The Gallup Presidential Approval Rating has for decades been the among the top barometers cited by media outlets measuring public opinion of the president’s performance.
President Trump has seen his rating by the agency slip in recent months, peaking at 47 percent last February and dipping to less than 37 percent in its last poll taken in December.
“This change is part of a broader, ongoing effort to align all of Gallup’s public work with its mission,” a spokesperson for Gallup said. “We look forward to continuing to offer independent research that adheres to the highest standards of social science.”
When asked by The Hill if Gallup had received any feedback from the White House or anyone in the current administration before making the decision, the spokesperson said, “this is a strategic shift solely based on Gallup’s research goals and priorities.”
Trump’s Gallup approval rating as of last December was among the lowest the organization had found since it began taking the poll in the 1930s.
The “strategic shift” argument makes no sense. At all.
Neither Gallup nor any other large survey research company conducts a standalone presidential job approval poll. It’s always part of a broader set of questions.
Judging from the RealClearPolitics aggregation of polls on the subject, the last time Gallup polled the question was a survey conducted December 1-15 of last year. Megan Brenan and Jeffrey M. Jones wrote it up as “Americans End Year in Gloomy Mood.” There’s not a link to the full survey, but the topics highlighted in their write-up included:
- Satisfaction with “the way things are going in the United States at this time”
- Americans’ assessment of current economic conditions
- What do you think is the most important problem facing this country today?
- Approval of Donald Trump, Fed Chair Jerome Powell, SecState Marco Rubio, and ten other high-ranking public officials
- Perception of Trump personal qualities (at least seven polled)
Judging from the news accounts, they’re still planning to ask the other questions, just not the ones about specific individuals.
Considering that presidential approval and “direction of the country” are the classic political polling questions, it strikes me as really odd for Gallup to suddenly stop asking the former. It costs them essentially nothing to keep asking it, and their numbers are guaranteed to generate huge amounts of reporting and analysis, all mentioning the Gallup poll.
Now, on the one hand, it’s not that big a loss. There is no shortage of polling on Trump job approval. Here’s the aforementioned RCP list:
This doesn’t even capture all the polls conducted since the start of the year, a mere six weeks ago.
And, frankly, Gallup isn’t even the most respected poll these days. Others have proven more accurate, at least for election forecasting.
Still, Gallup has the distinction of having been doing this much longer than others, giving us a way to compare over time. Indeed, the Hill report does just that:
Trump’s Gallup approval rating as of last December was among the lowest the organization had found since it began taking the poll in the 1930s.
Former President Truman earned an average approval rating of 45 percent during his time in office from April 1945 to January 1953; former President Biden earned an average approval rating of 42 percent from January 2021 to January 2025.
Former President Kennedy experienced some of the highest average ratings Gallup has ever recorded, 71 percent from January 1961 to November 1963, and former President Eisenhower topped out at an average of 61 percent from January 1953 to January 1961.
It’s a shame to lose this data point.
And again, the reason proffered makes no sense. This is the flagship political polling question and it costs essentially nothing to include it in any survey of public attitudes.
Indeed. For my generation Gallup is virtually synonymous with prez popularity, going back to Dewey’s defeat of Truman.
Sounds like time to invoke gVOR’s Second Law – if they won’t tell you why, it’s prudent to assume the worst.
When the reason offered by experts makes zero sense in their field, it’s not coincidence, it’s a clear lie.
As I said elsewhere:
‘If we print the truth, Trump will hurt us’.
‘If we print obvious false numbers, people will know that we are lying in our polls’.
(so let’s just hunker down)
Gallup stopped doing election polling after 2012, possibly due to their notoriously inaccurate final forecast, which showed Romney ahead in the popular vote by 1 point, when Obama went on to win by 4 points. (Rasmussem had the same final forecast, but unlike Gallup they never issued a mea culpa, nor did they stop doing election polling.)
Gallup has its problems and it has certainly been eclipsed in its historical role as the gold standard in polling (for a long time it was one of the only legitimate pollsters out there). But its approval center page, containing all their approval polls going back to Truman, is an invaluable resource. I hope they keep the page up.
If you don’t report bad numbers they don’t exist. That was Trump’s theory during Covid, right?
@Kylopod:
The people who monitor polling results by firm have long said that Rasmusson has a +5 point Republican bias. IIRC, some aggregators no longer include them because of that bias.
@Michael Cain:
No question. Curiously, though, during Biden’s presidency they were giving Biden some of his best approval numbers compared with other pollsters.