← Back to Archives

Did Diebold machines over-count for Schwarzenegger by a factor of 1.6?

It's getting late so I'm going to post this and run. I just got this in the inbox from the ever-vigilant Jody Holder. Apparently, (see the email below) if one looks at the relative amounts of votes cast in the California Recall election last fall over the different types of equipment used in the election, Diebold software seems to have counted 2.4 times as many Schwarzenegger votes as it should have. Of course, that statement is a tad suspect as the populations using a given set of equipment is in no way engineered to represent a common mixture of political persuasion (we'd expect different ratios of votes from different types of equipment). Not only that, but the math below is somewhat suspect (there's some sort of statistical weight being left out) and the email seems amateurish and the author appears with no credentials.

Oh well, since there is no permanent, tangible official record of votes cast on DRE equipment, we'll probably never know. That is, without something like a paper-based official record of each individual's vote, we can't audit the results of this election. In fact, if we had had such a record, we might even have caught the error (if there was one) while it was happening on election day (just like they did this past March in Napa County). Here's the text of the email I received... and here's a PDF of the below-mentioned Word file)

Friends:

I am not quite sure that you all understand the gravity of what the numbers below mean, so I shall present it from another angle:

  1. Schwarzenegger got 2.4 times as many votes as Bustamante from the Diebold machines on the average.

  2. Schwarzenegger got 6/10 of as many votes as Bustamante from all the other non-Diebold machines used. That is, on the average, Bustamante beat Schwarzenegger on all the other machines. 0.6 S/B + 2.4 S/B = 1.5 S/B where 1.5 is the State ratio of Schwarzenegger's votes to Bustamante's.

  3. The Diebold machines produced the victory for Schwarzenegger.

  4. IS THIS STATISTICALLY POSSIBLE WITHOUT THE MACHINES BEING PROGRAMMED?

  5. IS IT POSSIBLE FOR ONE MACHINE TO PRODUCE 1.6 TIMES AS MANY VOTES ON THE AVERAGE AS THE OTHER TYPE OF MACHINE WITHOUT HUMAN INTERVENTION? ( 2.4 / 1.5 = 1.6)

  6. It is entirely possible that Cruz Bustamante won the Recall election: Do you get it?

david

David Bayer [contact information deleted]

Posted by joebeone at Abril 5, 2004 12:34 AM | TrackBack