O'Reilly noted poll on effect of gay rights groups' endorsements, but not that he had been wrong about results

On The O'Reilly Factor, Bill O'Reilly read a letter criticizing him for “erroneously report[ing] that a poll found most Americans would not vote for a presidential candidate endorsed by a gay rights organization.” But while O'Reilly noted that the poll was taken “in a few states,” not nationally as he had earlier suggested, he did not acknowledge that his original assertion that the result applied to a “majority” of respondents was false.


During the August 15 edition of Fox News' The O'Reilly Factor, host Bill O'Reilly read an email from Cindi Creager of the Gay and Lesbian Alliance Against Defamation that criticized him for “erroneously report[ing] that a poll found most Americans would not vote for a presidential candidate endorsed by a gay rights organization.” Indeed, as Media Matters for America and the blog News Hounds noted, on the August 13 edition of the program, O'Reilly asserted that “a new poll” -- which he said was issued by “Pew Research or something like that” -- “says that most Americans won't vote for you if you get an endorsement by a gay rights group.” In fact, an August 6-8 Quinnipiac Poll of voters in Florida, Ohio, and Pennsylvania found that a majority of voters in each state responded that support for a presidential candidate by “gay rights groups” would not affect their level of support for the candidate. While O'Reilly noted that the poll was taken “in a few states,” not nationally as he had earlier suggested, he did not acknowledge that his original assertion that the result applied to a “majority” of respondents was false. Rather, he simply cited the Quinnipiac poll results from Florida -- which found that 28 percent of respondents would be “less likely” to support a candidate endorsed by a gay rights group, while 60 percent said it “would make no difference,” and 10 percent said it would make them “more likely” to support such a candidate -- and added, “That's what I was referring to.”

As Media Matters for America has documented, O'Reilly frequently misrepresents the results of polls, often either to cast the results more negatively for Democrats and progressives or more positively for Republicans. For instance, during the July 12, 2004, edition of his radio show, O'Reilly cited poll results showing that a majority of Americans opposed gay marriage to support his claim that “Far Left” senators were thwarting the will of the people by not passing a constitutional amendment banning gay marriage. In fact, numerous polls at the time showed that while the public did oppose same-sex marriage, a majority opposed amending the Constitution in order to ban it.

From the August 15 edition of Fox News' The O'Reilly Factor:

O'REILLY: Cindi Creager, Gay and Lesbian Alliance Against Defamation, New York City: “Bill, you erroneously reported that a poll found most Americans would not vote for a presidential candidate endorsed by a gay rights organization. We expect your program to do a better job living up to its no-spin promise.”

OK. Here is the Quinnipiac poll question that was asked in a few states, Florida being this example. “If a candidate is supported by a gay rights group, would you be more likely to vote for the candidate, less likely, or it doesn't make a difference?”

Ten percent, more likely; 28 percent, less likely; doesn't make a difference, 60 percent. That's what I was referring to.