> The theory is that a secret court order procedure on much broader surveillance program is enough to make NSA's surveillance acceptable to 27% of Democrats seems farfetched to me.
The procedural difference isn't the only difference in either the public facts or, more relevantly, the poll questions, and the issue isn't that it explains the entire difference, its that the difference in the circumstances and the poll questions makes the conclusion that the swing reflects a change of opinion on the same thing (whether partisan or otherwise) invalid.
The procedural difference isn't the only difference in either the public facts or, more relevantly, the poll questions, and the issue isn't that it explains the entire difference, its that the difference in the circumstances and the poll questions makes the conclusion that the swing reflects a change of opinion on the same thing (whether partisan or otherwise) invalid.