“Could Utah vote Democrat for president for first time in 5 decades?” Salt Lake City’s KUTV asked in an Aug. 1 headline. The CBS-affiliate certainly suggested it was possible, but it turns out its online article had misreported a two-month-old poll about the Hillary Clinton-Donald Trump matchup.
KUTV had reported that in Utah, Clinton was ahead of Trump by 1 point after the conventions. But, the poll it was reporting on was from June before the conventions and showed the two candidates tied.
In an e-mail to iMediaEthics, KUTV assistant news director Don Kauffman called the error “regrettable and preventable.”
“The mistake boils down to a bad hand-off between a broadcast reporter and our digital team,” Kauffman told iMediaEthics. “To prevent future mistakes we are emphasizing the importance of reporters writing their own web copy, and updating scripts to reflect the latest version of the story. ”
According to Kauffman, the broadcast video report was correct because the reporter “changed the wording in the audio booth” after realizing the poll in question wasn’t new. That reporter accurately reported that the poll wasn’t new and that both Clinton and Trump were tied with 35% in Utah. But, the change wasn’t made in that reporter’s script so the web writer who wrote the article for the KUTV website inaccurately reported it was a new poll. Then that web writer made a new error by typing Clinton had 36% instead of 35%.
“While writing the web text, the web editor either misread the script, or mistyped the number and indicated that Hilary Clinton’s percentage was 36%, instead of the correct 35%,” Kauffman told iMediaEthics. “It was a simple mistake, but combined with the earlier error referring to the poll as ‘new’ it led some readers to question the accuracy (or existence) of our source.”
KUTV posted a correction at the top of its article admitting the poll it cited was from June, not from after the conventions, and that the poll actually shows the candidates tied. The correction reads:
“Correction: This story has been updated to correct two errors. The Hinckley Institute-Salt Lake Tribune poll was first published in June 2016. It does not track data from immediately after the July conventions. KUTV also erred in showing Clinton with a 1-point lead over Trump. It has been corrected to reflect the Hinckley-Tribune poll, which showed each candidate at a statistical tie with 35 percent.”
The Washington Post‘s Philip Bump noted that even if the poll were current and did show Clinton ahead by a point, it doesn’t really mean anything for two reasons: “First of all, a 1-point lead is, of course, in the margin of error of the poll,” Bump noted. Further, the poll, as initially reported by KUTV, only accounts for just over 70% of the public so who knows where the rest of the votes would end up going.