[EM] 09/20/02 - The Manipulation Test:

[Elisabeth Varin/Stephane Rouillon]

Donald,

I do think IRV is better than Approval.
Mainly, because from an informatician point of view
using n ranks gives more information than only 2 or 3.
IRV's essential advantage is the easy
way it can get understood and the fact it is almost
identical to leadership runs (at least in Canada).

However, do not dig your own grave. Yes,
in a Condorcet method lower ranks can help defeat
our own first choice. I admit it.
But on that particular point, IRV is worst.
With IRV your first choice can make the difference and
make himself loose! I suppose you do not need any example.
If so we can provide...

All the residual approval methods I present on the
Electoral_systems_designers
site would elect A as IRV would with
60: A (yes) > B (yes)
40: B (yes) | A (no)
However, what should be pointed out is the degree of proportionality of
multiple winners methods when such a case gets repeated in every
riding...
IRV needs residual approval weights to optimize equity then.

Have a look,
http://www.fairvotecanada.org/phpBB/viewtopic.php?topic=8&forum=1&3
(see the fourth post)

Steph.


----
For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em