Public Policy Polling for Daily Kos & SEIU (3/17-20, MoE: ±3.1, registered voters, Obama trendlines 3/10-13, all others 3/4-6):
Last week, people were ready to offer all kinds of reasons (some more plausible than others) as to why the president's favorables had dropped six points and his job approval sank ten. This time, both measures are up six. So what gives?
I'm usually pretty reluctant to ascribe one-week fluctuations to anything other than normal churn, but this time, a major news event which directly implicates the United States has intervened. I'm speaking, of course, of the allied airstrikes on Libya - and those of you who were poll-watchers during the Bush regime will recall that Dubya almost always saw his poll numbers spike when he commenced (or renewed) military activity.
This time, though, there wasn't nearly as much buildup in the media as before Afghanistan or Iraq, and Obama didn't engage in the kind of bellicose saber-rattling that was almost the raison d'etre of the Cheney Administration. Moreover, the airstrikes began a little bit after we started our polling (we go into the field Friday through Sunday), so it's hard to tease out exactly when respondents would have been likely to learn of our involvement overseas. I would say, though, that it's not safe to assume all or even most respondents would have been aware of the attacks by the time they answered our poll—just as Daily Kos traffic drops on weekends, so does news consumption in general.
One possible hint that perhaps Libya did have something to do with it? Obama's job approval with men jumped 13 points... but that's an awfully thin tea leaf. What makes me particularly skeptical is that there's nothing out of the ordinary about this level of support from men—it's 44-51 right now, but back in early February, when Obama's overall job approval was an almost-identical 47-48, men gave him an almost-identical 44-52 rating. In other words, this number "fits," and if anything, last week's 38-58 among men was just unusually low.
So I want to wait at least a week before concluding that Libya is (or is not) the answer. These numbers are just too broadly similar to what we've seen in the past to come to firm conclusions - though perhaps we'll try to add a question on Libya for the coming week's poll and see if the crosstabs reveal anything. Suggestions for appropriate questions are more than welcome in comments!