Two recent polling scandals have intensified scrutiny of survey practices ahead of the 2026 elections. FiftyPlusOne says The Public Sentiment Institute admitted altering respondent answers in a Florida Republican primary poll commissioned by James Fishback, who finished third behind Jay Collins and Rep. Byron Donalds. Separately, a fake-poll operation called Median Strategies shut down after publishing fabricated results as a "social experiment," which briefly influenced online narratives in races including Los Angeles and Wisconsin. These incidents, combined with high-profile 2024 misses and reported large margins of error in recent primaries, have deepened concerns over polling reliability.
Second Pollster Admits Fabricating Data After Fake-Poll Scandal Shakes U.S. Races
Second polling firm admits to fabricating data after fake survey rocks LA mayor’s race
A second polling firm has acknowledged fabricating results in a 2026 primary contest, this time in Florida, after a polling watchdog investigated the company's work.
Investigators Find Altered Responses
Polling monitor FiftyPlusOne disclosed that it had been tipped off about The Public Sentiment Institute (TPSI) for failing to disclose that a recent survey of the Republican gubernatorial primary was commissioned by a candidate. Most major polling aggregators require firms to reveal when a campaign funds a survey because campaign-commissioned polls commonly favor the sponsoring candidate.
Byron Donalds won a commanding victory in the Florida gubernatorial primary (AP)
While probing that omission, FiftyPlusOne's analysts say they found a more serious problem: TPSI's own postmortem, published after Tuesday's Florida primary, admitted pollsters had altered some respondents' answers to benefit James Fishback, the candidate who commissioned the survey. Fishback ultimately finished third in the primary behind Jay Collins and the victor, Representative Byron Donalds.
"Jay Collins was polling in the twenties in our data. That did not align with our read of the race, and we concluded it might reflect a data-quality problem specific to our text-message panel. Acting on that conclusion, we identified respondents who selected Collins but whose broader profile aligned with a competing coalition, and leaned those respondents toward James Fishback," TPSI wrote.
FiftyPlusOne described that admission as an "obvious violation of standard polling practices," noting the issue was compounded by TPSI's failure to disclose its financial relationship with the Fishback campaign. The aggregator said it will exclude TPSI surveys from its analytics going forward.
A pollster commissioned by far-right candidate James Fishback did not disclose its backer and admitted later to faking some data (AP)
Related Fake-Poll Operation and Wider Implications
The TPSI disclosure comes days after another operation, Median Strategies, pulled down previously published results and shut down after admitting the polls were fabricated as a "social experiment." The operator, later identified by The Guardian as 21-year-old Rahil Prakash, said he wanted to test how easily fake polls could spread online.
Prakash's bogus polls briefly buoyed supporters of Francesca Hong before her defeat in Wisconsin's gubernatorial primary and were even cited by allies of Los Angeles Mayor Karen Bass during her re-election campaign, despite not being picked up by established polling aggregators.
LA Mayor Karen Bass touted a poll from 'Median Strategies' that ended up being a fake (Getty)
These incidents add to wider concerns about polling accuracy. The 2026 cycle follows notable polling errors in 2024 — including a high-profile Iowa poll that diverged sharply from the eventual result — and local analyses that have flagged large margins of error in recent primaries, such as a reported industry-wide error exceeding 10 points in a Michigan Democratic Senate primary.
Why it matters: Fabricated or altered polls and undisclosed campaign-funded surveys can distort public perception, influence media coverage and voter expectations, and further erode trust in election analytics at a moment when credibility is already fragile.
FiftyPlusOne's exclusion of TPSI and the collapse of Median Strategies underline growing pressure on pollsters and platforms to improve transparency, methodology, and verification to prevent fraudulent or misleading survey data from shaping political narratives.
Help us improve.
