Re: Arrow's Social Choice Theorem / Borda count

From: Ellis Wyatt (
Date: Thu Apr 27 2000 - 22:03:47 MDT

On Thu, 27 Apr 2000, Robin Hanson wrote:

> I never cared that much for Borda counting, but I do like approval
> voting, which is intuitive and easy to explain, and
> does seems to clearly improve on simple plurality.
> For listeners: approval voting is where each voter gets to vote for as many
> or few candidates as you want. The winners is the one with the most votes.

I like a modified form of elimination. We have the technology for it, of
course. Simple elimination is where all the candidates are ranked, then
in the first iteration, the candidate with the fewest #1 votes is
eliminated, and the other canditates "move up a notch" on the relevant
ballots. This is repeated as many times as there are candidates, minus
one. In a modified version, we could first eliminate outright anyone who
fell below a certain low average, and then maybe count #1, #2, and #3
votes for the first round or two in a large field.

Of course, it's going to be economically irrational for an individual to
vote, let alone spend time researching candidates, in any large election.
So, while the math/statistics here is intriguing, it doesn't make
democracy work.

--J. Goard--

This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:09:54 MDT