Gallup’s Presidential Approval Misstep

F or the first time since 1938, Gallup will no longer be asking its presidential approval question. Gallup’s rationale for dropping this question and its questions on political favorability after almost nine decades? The company said on Wednesday that the decision reflects its shift toward research on “issues and conditions that shape people’s lives.” But as a long-time polling watcher, I find Gallup’s decision disappointing for several reasons.

Let’s begin with Gallup’s perspective. In confirming the change, Gallup notes that there is no shortage of approval or favorability questions today from other pollsters. That’s true: In the past month, pollsters have asked more than a dozen presidential approval questions. And yes, Gallup has also changed its business model, and these questions don’t fit easily in their work on workplace culture and organization. While the Gallup website still includes the more familiar Gallup news releases on topics such as crime and abortion, they are no longer prominently featured on the site, having been eclipsed by the work Gallup is doing for its custom clients.

Why does it matter that we won’t see Gallup’s questions anymore? Today, many pollsters just throw their latest presidential approval rating into the air for a brief media hit, and they rarely provide the historical analysis Gallup provided regularly. So eager are some pollsters to give their audiences the latest finding on Donald Trump that some of them don’t even bother to compare their new findings to their own internal trends.

But Gallup gave us historical perspective. Here are a few recent examples of the kind of work the organization did on favorability and approval: In an October 2024 release, Gallup compared Trump and Kamala Harris’s favorability ratings to candidates going back decades, all the way to Eisenhower in 1956. In a 2025 release, the organization examined favorability ratings of six living presidents and included useful trends on each president’s favorability ratings over the course of their times in office. Another release looked at the gender gap in Trump’s approval ratings and compared it to the gap for other GOP figures. Will any other pollster give the kind of careful sustained attention to trends in the way Gallup has done with their long-running approval and favorability questions? At this point, none has the historical data that Gallup possesses.

This general neglect is part of a larger problem. Many pollsters these days show little interest in their own history, in the impressive work done by George Gallup, Elmo Roper, and others starting in the 1930s to take the public’s temperature on myriad issues. Pollsters’ partnerships with media organizations have narrowed much of polling’s focus to the latest news story. Don’t get me wrong: These questions are important, and pollsters should ask them. But there are other areas where a longer-term perspective is important.

Let’s take one familiar polling topic: the health of the American Dream. When formulating their surveys, few pollsters go back to the important work Gallup has done on the “ladder of life,” asking people to think about where they were on the steps of a ladder five (or ten) years ago, where they are today, or where they expect to be in five years. While the question doesn’t ask specifically about the Dream, their long-time trends provide a good baseline from which to evaluate contemporary concerns about it. It’s not my impression that many pollsters have bothered to look back to what I think may still be the most comprehensive survey ever on the subject of the American Dream, one done by the Roper Organization for the Wall Street Journal in 1985.

Questions asked in the past can inform the present. It’s not always possible to make direct comparisons, of course, as methodologies have changed, but consulting and updating old questions and trends provides a sense of what attitudinal anchors the public has had and may have today, in this case, about economic and social mobility.

With good polling, it takes time to understand what it is we are measuring. In today’s competitive polling environment, speed trumps understanding. Today’s pollsters also like to arrive at unambiguous conclusions, but polling often reveals tentativeness and ambiguity. When “don’t know” or “unsure” responses are in the 20–25 percent range — as they are on many topics today — it may be signaling that opinion is still forming or that the pollsters have asked a question in a way that doesn’t make sense to those taking the survey. Yet results are presented as firm.

Finally, as many polls have shown, the media have their own perspectives and biases, and polls today may reflect their preoccupations instead of the public’s. But by asking identical questions over time on presidential approval and favorability, Gallup has given the country a unique resource. Their decades-long commitment to these trends and many others reveals a public spiritedness that many pollsters and others lack.  Asking these questions over and over again and reporting the results in sober language no doubt enhanced Gallup brand and the trust people have had in it.

Gallup is still committed to several time series, including its invaluable Social Series, which gives us important trends in social, cultural, and governance areas. Other pollsters have been in the business for a long time now, and some of them update important trends regularly. At the time Gallup’s move was being discussed this week, the new British Social Attitudes Survey was released, documenting British attitudes yearly since 1983. Its long-time trends on immigration, taxes and spending, and first-past-the-post voting system were instructive, as was the serious over-time analysis provided by the survey’s principals.

Gallup’s choice to abandon its similarly educational and edifying long-term presidential approval and favorability trends is saddening, and its deep voice on these questions will certainly be missed. It’s a blow not just to the polling community, but to the general public as well.