It’s tempting to read this in conjunction with the Reuters poll from a few days ago that showed 49/41 approval for the ban and conclude that public support has begun to collapse in the days since. But that’s not what’s happening. It turns out the two polls were taken over the very same period, January 30-31. What we have here is a good old-fashioned poll conflict, especially in the numbers measuring disapproval.

Reuters phrased its version of the question differently, to specifically include the issue of refugees as well: “Do you agree or disagree with the Executive Order that President Trump signed blocking refugees and banning people from seven Muslim majority countries from entering the U.S.?” If it’s true that the sample in the Reuters poll is far more sympathetic to refugees than the sample in Gallup is, that might explain the overall difference in opinion on the order. But in reality, it seems to be the other way around: It’s the Gallup sample that strongly disapproves of suspending the refugee program whereas in Reuters there was only modest support (46/42) for the idea that the U.S. should open its borders to refugees of foreign conflicts.

The likelier explanation for the difference between Gallup is Reuters is the partisan split in the sample. Reuters actually included more Republicans (478) than Democrats (453), which isn’t necessarily inaccurate but is a bit unusual given the national population’s general Democratic skew. It may be that excitement over Trump as he takes office has led some fencesitters to identify as Republicans lately rather than independents. Gallup didn’t reveal the partisan split in its own sample, but it’s worth noting that the Republican/Democratic divide on the travel ban question is 83/14. With a spread like that, the partisan composition of the sample will matter a lot to determining overall public opinion. Before you dismiss the Gallup numbers on “too many Democrats!” grounds, though, do note: They have the travel ban at just 42 percent approval among independents as well.

On the other hand, here’s another new poll out today. This comes from YouGov, and was taken over a long time span than the Gallup or Reuters data — beginning on January 28th, the day after Trump issued the travel ban, until the 31st. Looks pretty solid:

He’s net positive on all four questions but solidly so on the 90-day ban for visitors. Is this another case, a la Reuters, where a ton of Republicans in the sample are driving the overall support numbers up? Uh, nope. Not at all. There are many more Democrats and independents in this poll than there are GOPers. The reason the 90-day visitor ban has solid overall support is because independents are firmly in favor and Republicans approve overwhelmingly:

Even a quarter of Democrats like the new proposal. Just to make all of this even more confusing and conflicted, although YouGov finds decent support here for Trump’s travel ban, they also find that his overall job approval has declined since he issued the order last Friday. On January 25, he was at 41/35; the order followed two days later, and then, on January 31, he was at 43/44. The slight downtick may have less to do with the travel ban, though, than with the general pace of events over the last two weeks. A friend emailed the other day when the travel-ban fight was raging and Supreme Court drama was peaking and there was some new media controversy or other burning up online and said, “It’s going to be like this every day for four years, isn’t it?” Yup, pretty much. Maybe some people are a little fatigued already.