This was really, really weird:
Frank Newport: Status Update on Gallup Election Polling Following Superstorm Sandy: Here's an update on Gallup’s thinking when we suspended our national daily tracking of the presidential election campaign as of Monday, Oct. 29. Basically, we reached the conclusion that Superstorm Sandy had compromised the ability of a national survey to provide a nationally representative assessment of the nation’s voting population…. [P]roblems… are concentrated in parts of the Middle Atlantic and Northeast and in specific areas within these regions. Thus, while we could have achieved the targeted number of completed interviews in these regions, those interviews would not have been representative of the overall population of the area…. [I]t is impossible to adequately weight to compensate for large segments of the population who cannot be reached at all in a survey, or in very low percentages, and whose opinions may have changed from previous, pre-storm measures.
Indeed. But you can weight the sample to compensate for large segments of the population who are unreachable *if you are willing to assume that their opinions have not changed from previous, pre-storm measures.
And a pollster that, say, believed in its likely voter screen would have done so.
It would have been much more honest for Gallup to have said on October 29: we are suspending our polling because we think there is something wrong with our likely voter screen.
Did the other tracking polls have an insurmountable Superstorm Sandy problem? No...
...And now Gallup is back, with its likely voter edge having moved 5% points toward Obama in a week that showed very little motion in any other poll...