« Is It All about White People with Kids? | Main | Whither the Labor Movement? »

What, If Anything, Do We Now Know about the Exit Polls That We Didnít Know Before?

As many readers of this blog probably already know, Scoop, a leftwing website out of New Zealand, has posted a large number of official documents that were issued by the NEP to subscribers on election night and the day after. These documents capture the exit poll results in three different stages: 3:59 and 7:33 pm on election day and 1:24 the next day.

What, if anything, do we now know from these documents that we didnít know before. Basically, nothing. What the documents do show is well-summarized by the indefatigable Mark Blumenthal over at Mystery Pollster:

On Election Day, the "national" exit poll had Kerry ahead by three points (51% to 48%) at 3:59 PM and by the same margin (51% to 48%) at 7:33 PM when the polls were closing on the east coast. By 1:33 PM the following day, the completed, weighted-to-match-the-vote exit poll showed Bush leading (51% to 48%). These numbers had been previously reported by the Washington Post's Richard Morin and Steve Coll on November 3.

The early samples included too many women: The percentage female fell from 58% at 3:59 PM to 54% at 7:33 PM, but this change alone did not alter the overall candidate standings (as the Simon/Baiman paper argues). By the next day, the sample was still 54% female, but the results among men and women were very different - Bush was 4 percentage points higher among men, 3 points higher among women.

All of the three releases are marked as "weighted," but keep in mind: The first two releases were weighted only to bring their geographical distribution into line with hard counts of actual turnout. The last release would have been weighted so that it matched the official count (something I explained here).

Keep in mind that the 7:33 PM sample from election night was incomplete. It had 11,027 interviews, but the next day NEP reported 13,660. The missing 2,633 interviews, presumably coming mostly from states in the Midwest and West, amounted to 19% of the complete sample.

We knew all this before, since this same basic story had been pieced together previously by various bloggers (including Blumenthal) and other observers from screen shots people saved, interviews with the NEP pollsters and other information about the exit polls that has been released since election night. But it is nice to have official documentation, as it were, that this story is correct.

Where does that leave us? I see no reason to change my previous position on this which I outlined in my post of November 12: exit polls have historically been off the actual election results before they are weighted to match those results and this year they were apparently off even more than normal. Why that was is an interesting question (see this post by Blumenthal for some of the logical possibilities), but there is no good reason to interpret this worsening of a long-standing problem as evidence of voter fraud--that is, that the unweighted-to-the-reported-election-result exits were right and the reported vote count was wrong.


Well, I think we also know that in this case at least the poll numbers which were not included in the second release were radically different: 19% of the poll sample caused a 6 percentage point swing.

We can impute that the remaining 19% preferred Bush over Kerry 64-35 (vs 48-51 among the 81%).

To me it's the result of a bad sampling plan: the first two results fully included east coast working people (9 to 5'rs anyway), but not midwest or western.

Ruy, can you then please explain the significance, if any, of this Washington Post exit poll of 13,047 respondents with MOE of 1% which shows Kerry handily leading Bush with women respondents at 54%. It is the Mitofsky NEP exit polling according to the explanation at the bottom of the graph, and is said to be the same poll that CNN briefly posted around 12:30AM on November 3rd.

Thank you.


Unless you live in Ukraine and your candidate is opposed by the west. Then such exit polls are proof of fraud. Also, if the unwieghted information is as useless as it apparently is, then why should we ever resoort for its even more suspect demographic estimates of the vote?!? Why don't we call a spade a spade...exit polls are completely unscientific.