Polling turns out to be a business in which generating data that confirm the thuddingly obvious can get you both money and news media attention.
Politico tried hard to make its story on a new poll sound exciting, choosing the headline, "Exclusive: GOP poll of women: Party 'stuck in past.'" Did we really need an exclusive, though, to find out that women in the Northeast are not especially fond of Republicans?
For that matter, did we need the poll itself to determine that Republican candidates should talk about other things after presenting their position on abortion, rather than just dwelling on it? And that women prefer candidates who profess concern about equal pay for equal work to candidates who dismiss that concern?
The Republican advocacy groups that commissioned the poll from a Republican strategy firm apparently thought so. But even the headline finding is trivial. An honest Democratic pollster, if he asked a focus group about various common criticisms of the Democratic Party, would find that a lot of people agree with some of those criticisms.
Some of the survey's other findings are just weird. The pollsters picked 11 conservative, or conservative-ish, policy ideas and asked people whether they would be more or less likely to vote for a candidate who supported them. For some reason, the pollsters concluded that support for charter schools and flextime doesn't "resonate" with women because those ideas did the worst among this set of options. Never mind that 56 percent of women still said that they would be "more likely" to vote for a candidate who wants to expand charter schools, and only 21 percent said they'd be less likely to do so.
Other polls, incidentally, have found pretty robust support for these ideas. A poll in June from a company run by the Republican strategist Kellyanne Conway found that 75 percent of women "strongly" favored flextime. It found plenty of female support for other conservative policy ideas, too.
I'm being tough on this poll, but the root problem isn't that the pollsters or the organizations that hired them are dumb. It's that they're thinking about the gender gap the wrong way -- and they're not alone. The entire political world, more or less, assumes that to win more elections Republicans have to address the distinctive problem they have with women. That assumption is easy to make, but it's wrong.
Women are slightly more liberal than men on issues of social welfare and war, and always have been. But gender influences voting much less than ideology, religiosity or marital status, as the survey itself confirms. And Republican candidates who win tend to have gender gaps about the same size as Republican candidates who lose. (George W. Bush did seven percentage points better among men than women in 2004, when he won, while John McCain's gap was only five points in 2008, when he lost.)
Maybe, then, Republicans should stop obsessing so much about women as a group. In recent years, voters have thought Democrats had more to offer the middle class than Republicans did. If that changed, Republicans would find themselves doing better among both men and women. They'd almost certainly still have a gender gap, but they'd be more likely to win elections.
The alternative path for Republicans is to keep paying pollsters to ask the wrong questions, and to keep getting answers that don't tell them anything useful.
To contact the writer of this article: Ramesh Ponnuru at email@example.com.
To contact the editor responsible for this column: Timothy Lavin at firstname.lastname@example.org.