More Bad Poll Data

This morning’s ABC News/Washington Post poll is getting a lot of press, with its apparently bad news for Republicans. The Post itself headlines its story “Filibuster Rule Change Opposed,” and begins its coverage of the poll with that issue:

As the Senate moves toward a major confrontation over judicial appointments, a strong majority of Americans oppose changing the rules to make it easier for Republican leaders to win confirmation of President Bush’s court nominees, according to the latest Washington Post-ABC News poll.
[B]y a 2 to 1 ratio, the public rejected easing Senate rules in a way that would make it harder for Democratic senators to prevent final action on Bush’s nominees.

Sounds bad. But here is the question the pollsters asked: “Would you support or oppose changing Senate rules to make it easier for the Republicans to confirm Bush’s judicial nominees?” That is an absurd question, to which I would probably answer “No,” too. The way the question is framed, it makes it sound like a one-way street, as though the Republicans wanted to change the rules to benefit only Republican nominees. If they asked a question like, “Do you think that if a majority of Senators support confirmation of a particular nominee, that nominee should be confirmed?” the percentages would probably reverse.
Of course, the poll contains bad news for Republicans across a broad range of issues, including Social Security. Which raises, as always, the question of the poll’s internals. Sure enough: they over-sampled Democrats. If you look at page 16 of the poll data, which can be downloaded from the Post’s article, it discloses that 35% of the poll’s respondents were Democrats, while only 28% were Republicans. Given that slightly more self-identified Republicans than Democrats voted in last November’s election, this represents an egregious, seven-point over-sampling of Democrats. No wonder the poll data are bad for Republicans.