Reply #102: Ida Briggs' original study is here:

[Febble]

http://www.invisibleida.com/New_Hampshire.htm

and her report on the recount is here:

http://www.invisibleida.com/index.html

My understanding is that selected precincts were chosen on the basis of her finding that vote-counted results were out of whack with expectations based on the 2000 vote, and that the degree to which they diverged was associated with voting technology. The implication from her reports is that the eleven precincts selected included those that she considered suspicious, but I cannot find a place where this is explicitly stated.

USAtoday has articles here:

http://www.russbaker.com/New%20Hampshire%20Recount,%20A...

and here:

http://www.usatoday.com/tech/news/techpolicy/evoting/20...

and there's a piece on Nader's website here:

http://www.votenader.org/media_press/index.php?cid=413

It sounds like Nader selected the 11 precincts based on Briggs's finding.

Ah, here's another one:

http://www.commondreams.org/headlines04/1118-03.htm

Precincts are described as those in which "Bush did surprisingly well".


On edit: I think there was a widespread belief, at the time of the NH recount, that the precincts were selected on the basis of aberrant exit polls, a trap Hertzgaard seems to have fallen into as well. Both the exit polls and the Briggs study pointed to the possibility of pro-Bush fraud, but the recount infirms the latter rather than the former. On the other hand, if one accepts (as one need not) that the hand recount is evidence against widespread electronic fraud, then the probability that the exit poll discrepancy was due to polling bias, rather than electronic vote-switching, at least increases.