Preliminary statistics indicate higher error for punch cards in recall race

See this preliminary analysis by Steven Hertzberg of Votewatch, a new non-partisan vote monitoring group. According to Hertzberg:

A preliminary analysis of the election returns shows that of the ballots cast, 42.5% were on Votomatic and Pollstar punch card machines. Our preliminary calculations show that Question #1 was either not marked by the voter or recorded by the equipment in 7.7% of the ballots cast on these machines. The average "not counted/marked" rate for the remaining voting systems is 2.3%, with the next highest rate being the Optech optical scanner at 4.35%."

Of course, a full study will have to control for other factors. For example, voters in counties with punch cards might have been more or less likely for other reasons to have abstained on question 1. But the preliminary figures raise reasons for further concern over the use of such machines.
UPDATE: See also this A.P. report.

Posted by Rick Hasen at October 10, 2003 10:28 AM