User:Cymru/Independence of correlated candidates criterion: Difference between revisions

Content added Content deleted

Inline

Revision as of 23:16, 20 September 2005

The Independence of Equivalent Candidates criterion was devised in 2005 by Thomas Smith. It was inspired by the inadequate definition of the independence of clones criterion which does not properly address rated voting systems and presumes that equivalent candidates are identical or very similar to each other.

Equivalent candidates are a group of candidates that are supportively ranked in adjacent positions, where equal ranking is not allowed, or equally ranked or rated, where equality is allowed, on every cast ballot. Thus, an approved and a disapproved candidate can never be "clones" as they are not rated equivalently.

A voting method is said to be independent of equivalent candidates if the addition or removal of a equivalent candidate from a particular set of equivalent candidates in the tally, would only change the winner to another member of the set, if the original winner was a member of that set.

Methods that satisfy this criterion are [[approval voting], range voting, Bucklin voting, Schulze method, and Instant-runoff voting.

Independence of equivalent candidates is a weakened form of the independence of irrelevant alternatives criterion.

@@ Line 1: / Line 1: @@
-The '''Independence of Correlated Candidates criterion''' was devised in 2005 by Thomas Smith.  It was inspired by the inadequate definition of the [[Strategic nomination|independence of clones]] criterion which does not properly address rated voting systems and presumes that correlated candidates are identical or very similar to each other.
+The '''Independence of Equivalent Candidates criterion''' was devised in 2005 by Thomas Smith.  It was inspired by the inadequate definition of the [[Strategic nomination|independence of clones]] criterion which does not properly address rated voting systems and presumes that equivalent candidates are identical or very similar to each other.
-Correlated candidates are a group of candidates that are supportively ranked or rated in adjacent or equal positions, on every cast ballot.  This definition excludes candidates who are differentiated by disapproval or 0 rating from being correlated with those given supportive ranking or rating.
+Equivalent candidates are a group of candidates that are supportively ranked in adjacent positions, where equal ranking is not allowed, or equally ranked or rated, where equality is allowed, on every cast ballot.  Thus, an approved and a disapproved candidate can never be "clones" as they are not rated equivalently.
-A voting method is said to be independent of correlated candidates if the addition or removal of a correlated candidate from a particular set of correlated candidates in the tally, would only change the winner to another member of the set, if the original winner was a member of that set.
+A voting method is said to be independent of equivalent candidates if the addition or removal of a equivalent candidate from a particular set of equivalent candidates in the tally, would only change the winner to another member of the set, if the original winner was a member of that set.
+Methods that satisfy this criterion are [[approval voting], [[range voting]], [[Bucklin voting]], [[Schulze method]], and [[Instant-runoff voting]].
-Independence of correlated candidates is a weakened form of the [[independence of irrelevant alternatives]] criterion.
+Independence of equivalent candidates is a weakened form of the [[independence of irrelevant alternatives]] criterion.