Nate Silver tweets these three things:
bq. Politico: Obama’s approval rating down 5 points with Jewish voters since June. (Unmentioned: it’s also down 5 points among ALL voters.)
bq. Basically, the Politico article presents objective evidence that Obama does not have a “Jewish problem”. Author concludes the opposite.
bq. Also might be worth mentioning: a 5% approval decline among Jewish voters = a 0.1% decline in Obama’s overall rating.
It’s worth remembering that this exact same narrative emerged in 2008. Here is a NY Times article from May 22, 2008, with the headline “As Obama Heads to Florida, Many of Its Jews Have Doubts.” (The article also falsely asserted — without citing evidence — that Jews were generally “drifting to the right.” See my old post documenting that the percentage of Jews voting Democratic has increased over time. As Jon Cohen points out, Obama did as well or better among Jewish voters than did Clinton or Kerry.)
So why won’t this meme die? In general, intra-party strife is newsworthy, as I’ve noted before. It’s not interesting when evangelical Christians dislike Obama. But when the base, including most Jewish voters, is “angry” or “demobilized” — no matter how weak the evidence — a thousand stories are spawned.
Commentators also tend to underestimate how much a campaign rallies the base. Fourteen months out, things look bad for Obama and Democrats are fretful. Their angst makes headlines. But no one bothers to remember that campaigns really do bring wayward partisans home. This is the most venerable finding from political science studies of campaigns. I wrote about it here.
But there is a fundamental problem in the interpretation of polls as well, which Silver gets at. When analyzing polls, you must always ask: “compared to what?” Very little in polls is definitively large or small in absolute terms. It’s the relative comparisons that matter. A 5-point decline among Jews says little about Jews if there’s been a 5-point decline among basically every other demographic group.
The fixation on a trend among one group is doubly misleading because it gets your mind thinking about explanations idiosyncratic to that group. So with Jews, it’s because of Obama’s alleged dovishness on Israel. With Latinos, it’s because he hasn’t pushed comprehensive immigration reform. With working-class whites, it’s because he’s too elitist. And so on.
What’s happened to Obama is not a “Jewish problem.” It’s an “economy problem.”