To: Selectric II who wrote (117968 ) 12/19/2000 10:54:43 PM From: mst2000 Read Replies (1) | Respond to The recovery rate statistic has been written about extensively -- the notion is that, within a relatively narrow range, about the same percentage of people in each county throughout the nation go to the voting booth, but don't vote for President. To the extent counties show "undervotes" at a materially higher rate (say, five times as high, as was the case in these 3 counties, and 1 or 2 others in Florida), it is fair to conclude that it is due to some extent to the failure of voting machinery to record votes properly (something notoriously true with punch ballots, as admitted by the inventor of the machine during cross-examination on the witness stand in this trial). It has been shown not to be a matter of demographics, although the education level of voters does have an affect (and may have been responsible for the higher level of "voter error" in these particular counties). A proper recovery rate tends to bring the undervote in these counties more in line with the undervote in counties with more accurate voting machinery (like optical scanners), but not necessarily 100% in line because of the education and demographics. A manual recount is intended solely to recover those votes that can be discerned by the naked eye, but not by the machine. The pregnant and perforated chads do indicate voter intent, albeit imperfectly. But clearly, it is not "voodoo democratic mindreading" and it is not mere coincidence that the undervote rates are so high in counties with punch ballots.
