David Reel: It’s hard to poll an electorate that doesn’t trust pollsters
In 1935, George Gallup left his position as a college professor to launch the American Institute of Public Opinion, now known as the Gallup Organization. His initial venture into polling had mixed success.
His prediction that Franklin Roosevelt would win in the 1936 presidential election was correct. He predicted that Harry Truman would lose in the 1948 presidential election bid and was wrong. Gallup acknowledged his mistake was due to his decision to suspend polling three weeks before election day.
Despite that spectacular failure in 1936, the Gallup Organization has been hugely successful ever since then. Their polling results have been and are published regularly in newspapers in the U.S. and abroad.
In 1937, Gallup launched public polling on job performance ratings on incumbent U.S. presidents. Last December, Gallup announced they will stop doing those polls.
Instead, they will focus on broader issue-based independent analytics and policy research. A Gallup spokesperson has said this business pivot was not in response to pressure from the Trump administration. The spokesperson said the shift was “solely based on Gallup’s research goals and priorities.”
While Gallup was a prime source of presidential job performance approval ratings, and will be missed by the media, candidates, campaign consultants, think tanks, and political pundits, there will be no shortage of polling firms offering comparable information.
The polling results from all these firms will be analyzed by political candidates to develop messages for distribution to voters and influence their voting behavior in a wide range of elections, most notably the 2026 midterm elections for the U.S. Congress, especially for the 435 seats in the House of Representatives.
For the Democratic Party, this election will be an opportunity to win enough seats to reclaim a majority in the U.S. House for at least the next two years, while for the Republican Party will be an opportunity to win enough seats to maintain or expand their current House majority for at least the next two years.
Both parties are well aware that after some previous midterm elections, there were dramatic flips in which party controlled the House.
In 1994, with an incumbent Democratic president in office, 54 House seats flipped from blue (Democratic) to red (Republican). In 2006, with an incumbent Republican president in office, 31 House seats flipped from red to blue. In 2010, with an incumbent Democratic president in office, 63 House seats flipped from blue to red. In 2018, with an incumbent Republican president in office, 41 House seats flipped from red to blue.
These results have led some political observers to conclude that a significant number of voters who are not satisfied with the job performance of an incumbent president, will choose to vote for a congressional candidate of the opposite party.
Some political observers are also predicting outcomes for the 2026 midterms will be comparable to the midterm flip elections in 1994, 2006, 2010, and 2018.
They base their suggestions on regular media reports that only 40 percent of voters approve of Donald Trump’s job performance on a wide range of issues including immigration, inflation, food prices, gas prices, prescription drug prices, and overall economic growth.
However, some political realities need to be acknowledged with regard to those media reports and others who track all the issues of concern to all the votes.
In today’s deeply divided society, some voters are reluctant or even unwilling to respond to pollsters even if told their answers on which candidates they support, their future voting plans, positions on select public policy issues, points of view on candidates, and incumbent office holders are confidential.
We live in a VUCA world — a world with high levels of volatility, uncertainty, complexity and ambiguity. Voters are often fickle, sometimes irrational, and never fully predictable.
The November midterm elections are almost ten months from now. Between now and then, much can happen resulting in dramatic shifts in voters’ views on President Trump’s performance as well his positions on current and emerging issues.
A recent Gallup poll reported that most Americans now actually have high positive expectations about the economy for the next six months, despite news from other polls reporting lingering negative expectations. This pool also reported more Americans are positive rather than negative about the future of the stock market and economic growth, likely based on recent news the Dow Jones Industrial Average closed above 50,000 for the first time ever.
The rate of inflation cooled to 2.4 percent over the 12 months ending in January, with the lowest rise in the consumer price index since May 2025.
Last month, grocery prices also increased by the smallest amount since last summer.
The economy added more jobs than expected in January, with the national unemployment rate dropping to 4.3 percent, an early sign that the labor market may be regaining strength.
The most current information, both positive and negative, on every issue of concern to voters is available 24/7/365 to help them make informed decisions on those issues.
Only after returns are certified in November can any meaningful evaluation be done on if, when, and how incumbent presidents’ job approval ratings impact midterm election voting.
David Reel is a public affairs/public relations consultant who serves as a trusted advisor on strategy, advocacy and media. He was formerly active in the government and political arenas in Harrisburg and Philadelphia. He now lives and works from Easton on Maryland’s Eastern Shore.
