Ranked Choice Voting -- San Francisco Style

From: <dr-jekyll_at_att_dot_net>
Date: Tue May 04 2004 - 22:37:48 CDT

I called the San Francisco Department of Elections and talked with someone named Anita. While she doesn't have all the details, what she did have enabled me to update the Data Model I proposed and develop some suggestions for the Process Model (or Programming Specs or whatever we'll call it).

The Data Model changes were easy.

1) Add one data item, Ranked_Choice_Lim, to the contest definition table (CFO_Cont). Each contest that allows Ranked Choice Voting will have a number greater than one, which is the maximum number of ranked choices per ballot for this contest.

2) Add one data item, Ranked_Choice_Num, to the Ballot Segment for Candidate For Office (BS_CFO).

The Process Model is complicated considerably more by this. Here are my suggestions:

1) Materialize Ballot on Screen (Or whatever we'll call it) will need to allow for ranked choice numbers rather just check marks or x's for voters' choices.

2) Checking for overvotes and undervotes will need to refer to the ranked choice information.

3) Counting Votes By Contest -- With ranked choice voting, count only those ballots with Ranked_Choice_Num = 1.

4) Finding Ranked Choice Majority -- If no candidate wins a majority, copy all votes for this contest to a working area and switch to decision support logic. The candidate with the lowest vote total is to be eliminated. (What if tie for last -- TBD).

5) Eliminate a Ranked Choice Candidate -- To eliminate a candidate, delete all his 1st choice votes, mark the #2 choices on those ballots as #1, and cascade this logic to subsequent choices (3 becomes 2, 4 becomes 3, etc). Count votes with Ranked_Choice_Num = 1 and check for majority. If still no majority, again identify lowest vote getter and repeat the process.

I hope this helps. If anyone else knows of anything else like this, let me know. If there are rules, I can model it.

Kurt
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Mon May 31 23:17:13 2004

This archive was generated by hypermail 2.1.8 : Mon May 31 2004 - 23:18:15 CDT