Human error explains votes that ‘disappeared’ in CNN coverage of California recall
By Ciara O'Rourke
CNN among other news organizations projected that California Gov. Gavin Newsom will hold onto his job after an election to decide whether he should be recalled.
According to unofficial election results posted on California’s secretary of state website, nearly 64% of voters opposed recalling Newsom compared with about 36% who voted yes.
Those numbers will change before the results are certified on Oct. 22 as more mail-in and provisional ballots are tallied, but Newsom’s lead wasn’t in question.
Still, unfounded claims of election fraud loomed as people headed to the polls on Sept. 14.
In one video from that evening that’s being shared on social media, a man recorded his television, which was tuned to CNN.
The screen shows a graphic that says: "Should Gov. Newsom be recalled?" The graphic then reflected that 4,530,002 "no" votes had been counted — 67.1% of the votes tallied by that point — and 2,225,915 "yes" votes had been counted, or 32.9%. In the bottom right corner of the graphic it says "52% estimated vote," meaning that these totals only account for about half of all ballots cast in the race.
The man recording the screen then zooms in on a similar, smaller graphic with the same information in the bottom right corner of the screen.
"2.25 — watch," he says. The graphic then changes and some of the numbers change. The "yes" votes go from 2,225,915 to 1,874,206, and from 32.9% of the vote to 29.3% of the vote. The number of "no" votes, meanwhile, don’t change, and the percentage of "no" votes increases to 70.7%. The estimated vote changed from 52% to 49%.
"1.874," the man says. "Where did they go? 400,000 votes just disappeared? Wow."
The man wasn’t seeing things, and the recording of the CNN broadcast is authentic. But the polling firm that provides election data to CNN and other news organizations told us that what that man witnessed was a mistake being corrected.
This post was flagged as part of Facebook’s efforts to combat false news and misinformation on its News Feed.
CNN declined to comment for this story. But it got those numbers from Edison Research, a polling firm that provides elections data to CNN among other news organizations.
Rob Farbman, the firm’s executive vice president, told PolitiFact that Edison relies on different sources including state data and reporters hired to collect vote tallies in person at county elections offices.
A reporter hired by Edison who was stationed at an office in Santa Clara County called the firm and accidentally said that about 350,000 votes there were in favor of recalling Newsom. In reality, that number reflected all tallied votes in that county, no and yes votes combined. The mistake resulted in CNN wrongly reporting that there were about 350,000 extra votes to recall Newsom.
"This error was entered at 11:19pm ET and corrected 2 minutes later at 11:21pm ET," Farbman said in an email. The vote ‘drop’ was taking out the error and putting in the vote being reported by Santa Clara County."
The estimated vote changed from 52% to 49% because when the correction was made, "the total number of votes tabulated declined by 350,000 votes and estimated percent of the total vote currently tabulated declined," he said.
But those votes weren’t surreptitiously scrubbed from the election. They were mistakenly entered, and then removed.
We rate claims that these votes disappeared fraudulently False.
----------------------------------------------------------------------------------------------