Gallup, Up, Up, and Away: How Presidential "Approval" Has Been Polling's Most Misleading Question

"In general, do you approve or disapprove of the way [president's name] is handling his job as president?" 

Chances are, unless you've been living under a rock these past two decades, you've seen this phrase at some point in relation to public opinion. This phrase was coined by Gallup, an American public opinion polling agency that first began asking a version of that question in 1938, to give the American people living through the Great Depression an opportunity to share their thoughts at a time when they had no other opportunity to do so. This was well before the now tiresome trope of "talking to voters in diners," and for decades was the only way for an average American to share his or her politics in a way that would make national news. In the early decades of Gallup, the agency would send out individuals and talk to people at their doors, a remarkable undertaking during a time when the internet, home phones, and even reliable public transportation did not exist. Gallup was a pioneer in providing a snapshot in time as it related to the American president. Newspapers would pay good money to access the results of the Gallup polls to pass along to their audiences. While not a perfect predictor (Gallup notably was part of the "Dewey Defeats Truman" groupthink of 1948), the agency by and large was seen as the gold standard up through the early years of the 21st century.

But in 2012, that all changed. Gallup's final presidential poll predicted a one-point Mitt Romney win. When it ended up being a four-point Barack Obama victory, the agency reevaluated its methods and ultimately decided to forgo doing head-to-head polling for the 2016 presidential election. This announcement marked a new era, one where the advanced analytics of people like Nate Silver were now at the forefront. After having correctly predicted the 2012 presidential results in all 50 states, Silver and his team at 538 were now seen as the gold standard, replacing more traditional pollsters like Gallup. Aggregation was the key, and by bringing together data from countless polls and assigning them a weight based on their methodology, Silver felt he had cracked the code to 21st-century presidential polling. Of course, 2016 quickly dethroned 538 as the top dog when even their advanced analytics missed out on Donald Trump's victory. Silver tried to claim credit for his team's work, but being the best, bad pollster rang hollow to an American public that was quickly losing confidence in what was feeling more and more like outdated junk science rather than an accurate snapshot of the American electorate. 

With Gallup out of the head-to-head presidential matchup game, it could instead focus on its polls measuring favorability. Over the past ten years, Gallup continued to measure presidential favorability as a way to try to identify broader public trends. Gallup was not the only pollster to measure this, as organizations like Fox, CNN, and NBC also polled the American people on this subject. This past November's Gallup poll showed Trump at 36% approval, a new low for his second term and trending downward toward his all-time low of 31% reached in January of 2021. Gallup has also continued to survey the American people on other aspects of their lives, and their most recent poll has shown American optimism to be at an all-time low. It doesn't take a political scientist to put two and two together; Americans are pessimistic, and the reason for that pessimism is clearly Donald Trump. Continuing on the path we're on is likely to result in Trump surpassing his all-time low approval rating, another huge blow to this administration before the 2026 midterms. 

But rather than stay the course, Gallup has chosen to remove itself from the game. From The Guardian: 
 

Gallup, the public opinion polling agency, will stop tracking presidential approval ratings after almost nine decades, a spokesperson confirmed.

As Donald Trump continues to closely scrutinize polling of his popularity, and publicly lambast media companies that report on unfavorable numbers, Gallup insisted its decision was “solely based on Gallup’s research goals and priorities”.

The company said on Wednesday that it would stop measuring the favorability rating of individual political figures, which “reflects an evolution in how Gallup focuses its public research and thought leadership”, after 88 years.

“Our commitment is to long-term, methodologically sound research on issues and conditions that shape people’s lives,” a spokesperson for the agency told multiple media outlets. “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.”

The Gallup Presidential Approval Rating is one of the most cited barometers of public opinion on the president’s performance, and has been in use since Harry Truman was in office. The polling provides a snapshot of American history through its data, including a whopping 90% approval rating for George W Bush in the days following the 9/11 attacks.

Trump’s second-term approval rating fell to 36% in December, which is among the lowest ever recorded by the company. The US president started his second term with an approval rating of 47%.

Thus, Gallup ends its 88-year polling question with a whisper rather than a bang. 

But we can't help but question the timing of their decision. With Trump's steadily declining numbers, did Gallup feel pressured to end their longstanding presidential approval question? Did they want to avoid a potential lawsuit like the one Donald Trump brought against longtime Iowa pollster Ann Selzer? Or is this simply Gallup's recognition that capturing a president's approval rating is no longer a best practice in the 21st century? 

Whatever the reason, we should all be grateful that Gallup is no longer asking this particular question. For too long, too much media attention was created through Gallup's presidential approval polls. For nearly 90 years, we've been relying on "vibes" rather than measurable outcomes of presidential administrations. Polling is done as a way to fill the void during the time when there aren't active presidential campaigns. Presidential approval rating has always been a lazy and unscientific way to measure the effectiveness of a president and his policies. We saw Joe Biden's approval rating take a nosedive in 2021, not because of what he was doing during the Afghanistan withdrawal, but how he was perceived for doing it. The inherent flaw in polling is that participants only pick up the phone if they already have strong feelings about a person or topic. With the news media intentionally skewing its Afghanistan withdrawal coverage to make Joe Biden look bad for ending their forever war, it should have come as no surprise that his Gallup approval rating dipped 14 points over the summer of that year. Joe Biden never received credit for the largest evacuation of an active warzone since Dunkirk; instead, he was criticized nonstop by the media, and that portrayal clearly impacted any potential poll takers that Gallup contacted over those summer months. Because of this singular incident, Joe Biden's presidential approval rating never recovered. 

Nowhere else do we unjustly correlate public opinion with job performance. Imagine if actors were polled on whether the public approved of their acting. Or imagine if musical artists were polled on whether fans approved of their latest album or concert tour. Imagine if the public were polled as to whether or not they approved of the play of a particular athlete. In these fields, you are judged by your production alone: how many people see your movies, download your album, or buy your jerseys. Sure, people have opinions about these individuals. Some even have strong opinions about certain celebrities. But those opinions are confined largely to social media and nowhere else. A performer's ability to continue to do their work is never adversely affected by a snapshot in time when a random assortment of people are asked to evaluate their performance based on vibes and vibes alone. 

And that's why we should have never taken Gallup seriously. Polls aren't news; they're filler. They're done as a way to keep a presidential election in the news, 3+ years away from that election taking place. Even in the workplace, if you are asked if you approve of the way someone is handling their job, that question is never asked in an isolated way. Instead, there are follow-up questions to generate clarity on your response. Gallup never offered that. You simply said yes or no if you approve of the way the president was handling his job. Odds are, your response would be the same no matter at what point of the presidency you were asked this question. Yet for nearly 90 years, we took Gallup as gospel. We made entire presidencies rise and fall based on vibes, based on whether or not we approved of the job the president was doing. For 90 years, we had to live with Gallup creating a narrative when none existed.

Polling has always been an imperfect science at best. Since 2016, we've seen polls miss time and time again. Presidential elections. Midterm elections. Special elections. Technology has outpaced what has become an antiquated practice. People block calls from unknown numbers. Younger generations no longer have landlines. Those who actually answer and respond to a 15-minute automated survey have the time and resources others simply do not have. We'll continue to see polling in 2026 and 2028. We'll continue to see dissatisfaction with Donald Trump and the Republican Party. But while Gallup may no longer measure the false vibes its presidential poll question generated over 88 years, we simply cannot deny that this is a historically unpopular president serving at the helm of a historically unpopular political party. People don't need to answer a phone survey to tell us that. We see it through marches, protests, and community meetings. We even see it (in minuscule amounts) from the party in power itself, with a small but growing number of Republicans openly tiring of the Trump Imperial Presidency. This visible disdain for Donald Trump tells a much better story of the lack of "approval" so much better than any poll ever could. 

Because not even Gallup can measure the true number of people who utterly despise Donald Trump.