Exit polling means catching voters when they're leaving the polls. The samples are much larger and more carefully planned than normal polls. So far we only have normal polls which happen to ask whether people have already voted or not.
assuming that you conducted an exit poll of those who voted early, you could only make valid projections if you understood the demographics of those who voted ...
for example, let's say (hypothetically) that 80% of those voting early in a given state were Democrats ... and let's say your exit poll showed Kerry winning 70% of the early vote ... you might not have enough information to determine what the rest of the voters will do ...
i'm somewhat inclined to ignore exit polls of early voters this year ... it seems like there are too many variables ...
Typically, republicans vote by mail in higher number than democrats. And it seems that at least this year democrats vote in person early in higher numbers than republicans. See for example the thread about NH. A poll shows Kerry 12 pts ahead among people who have already voted, but tied among those who will vote on election day.
at least in Florida the last time around. Unfortunately, based on that precedent, you can't rely on them, since you can't be sure the person's vote was counted as intended!
Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators
Important Notices: By participating on this discussion
board, visitors agree to abide by the rules outlined on our Rules
page. Messages posted on the Democratic Underground Discussion Forums are the
opinions of the individuals who post them, and do not necessarily represent
the opinions of Democratic Underground, LLC.