March 8, 2013
President Barack Obama speaks about the sequester. (Charles Dharapak/Associated Press) President Barack Obama speaks about the sequester. (Charles Dharapak/Associated Press)

Great item from Sean Sullivan this afternoon looking at the emerging conventional wisdom that Barack Obama’s polling numbers have tanked, and concluding: too early to tell.

Sullivan gets this exactly right: “For starters, we need to see more post-sequestration polling numbers (the cuts kicked into effect last weekend) to get a better sense of what the cuts ultimately mean for Obama’s political standing.” Indeed, the HuffPollster approval model shows Obama slipping less than a single percentage point from his December peak — although that depends on how sensitive the model is asked to be to change.

As always with polling, it’s much better to look at polling averages than any single result. But unlike the weeks before an election, few pollsters are reporting results during March of a non-election. The polling story is mainly being driven by Gallup, which not only has long-term problems of its own but also is, as all polls are, vulnerable to all sorts of random variation. Obama’s Gallup approval rating does appear to be down a few points over the last ten days or so … but it’s still very possible that it’s still just random noise.

There’s also a Quinnipiac survey showing Obama with a low 45 percent approval rating, below even Gallup. However, the previous Q poll, a month earlier, had Obama at 46 percent. Not a lot of change there!

It’s certainly possible that Obama’s approval ratings have in fact declined a bit. It’s also possible that they haven’t — or that they’ve declined even more than the polling we’ve seen so far suggests. It’s just impossible to tell much. Polling averages are the first step — but when there’s relatively little polling, even the averages need to be used with a lot of caution.    It just takes more information to find out what’s going on.

Continue reading