O'Reilly misleading on same-sex marriage ... againJuly 16, 2004 4:58 PM EDT ››› GABE WILDAU
On the same day that Media Matters for America exposed FOX News Channel host Bill O'Reilly distorting same-sex marriage poll data on the July 12 edition of The O'Reilly Factor, he misled viewers on the issue again.
O'Reilly billed the "Talking Points Memo" segment of his July 15 show as "[t]he whole truth about same-sex marriage"; yet seconds later, he said the following: "As far as the constitutional amendment is concerned, most polls say more than half the nation wants man-woman marriage to have constitutional protection, but it will not happen because the Senate will not approve it and that's that."
O'Reilly's statement was misleading because it addressed only those polls that present the question of a constitutional amendment on marriage in a particular way. The truth is that while polls show that Americans largely oppose same-sex marriage, they also indicate that if the question offers the alternative of deciding the issue at the state level, the public prefers that over a federal constitutional amendment. As Media Matters for America previously explained, a survey (pdf) by the Annenberg Public Policy Center of the University of Pennsylvania from late June indicates that 48 percent of Americans oppose "an amendment to the U.S. Constitution saying that no state can allow two men or two women to marry each other," while only 43 percent are in favor. An ABC News/Washington Post poll (pdf) from early March showed 53 percent in favor of letting states decide the question, compared with 44 percent in favor of an amendment.
A minority of polls (see a roundup of contrasting polls here [no subscription required] or a more complete compilation here [NationalJournal.com subscription required]) appear to show the public favoring a constitutional amendment; however, as ABC News explained in an analysis accompanying its January survey, poll results on this issue are unusually sensitive to the precise wording of the question. In general, respondents are less likely to say they favor an amendment if the questioner offers the alternative option of letting states decide the issue; they are more likely to favor the amendment when the questioner describes the proposed amendment as defining marriage as a union between a man and a woman -- but less likely again if they are explicitly informed that the effect would be to legally ban same-sex marriage.
For example, a CBS/New York Times poll (pdf) from March showed the public appearing to favor a constitutional amendment; yet the same poll found that a narrow plurality of Americans believes that "laws regarding marriages and civil unions" should be determined by state governments rather than the federal government. The same poll also showed that only 38 percent of the public believes that "defining marriage as a union only between a man and a woman is an important enough issue to be worth changing the Constitution for," compared with 56 percent who believe it's "not that kind of issue."