To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.admin.generalOpen lugnet.admin.general in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Administrative / General / 2565
2564  |  2566
Subject: 
Boolean vs. multivalued votes
Newsgroups: 
lugnet.admin.general, lugnet.off-topic.geek
Followup-To: 
lugnet.off-topic.geek
Date: 
Tue, 17 Aug 1999 22:43:28 GMT
Viewed: 
240 times
  
I agree 100% with what Larry wrote here:

   http://www.lugnet.com/cad/dev/?n=2672

On the larger subject of voting, however, I've always felt that up/down is
too coarse...that -N/+N systems are more expressive.

For example, N=3...

   Proposition:  Blah blah blah blah.

   Vote:  -3  Disagree strongly
          -2  Disagree
          -1  Disagree somewhat
           0  Ambivalent
          +1  Agree somewhat
          +2  Agree
          +3  Agree strongly

   An average of      is similar to
   -------------   -------------------
        0.0            a tie    (50%)
        1.0        2/3 majority (~67%)
        1.5        3/4 majority (75%)
        2.0        5/6 majority (~83%)

Or N=2...

   Proposition:  Blah blah blah blah.

   Vote:  -2  Disagree strongly
          -1  Disagree
           0  Ambivalent
          +1  Agree
          +2  Agree strongly

   An average of      is similar to
   -------------   -------------------
        0.0            a tie    (50%)
        0.667      2/3 majority (~67%)
        1.0        3/4 majority (75%)

Or N=1...

   Proposition:  Blah blah blah blah.

   Vote:  -1  Disagree
           0  Ambivalent
          +1  Agree

   An average of      is similar to
   -------------   -------------------
        0.0            a tie    (50%)
        0.333      2/3 majority (~67%)
        0.5        3/4 majority (75%)

Multivalued votes allow for people with mild opinions to vote "kinda yes" or
"kinda no" and for people with very strong opinions to vote "way yes" or
"way no."  This is important to me because everyone has different levels of
understanding and expertise.  In a boolean (thumbs up or thumbs down) voting
system, someone who agrees mildly with something might vote "no" on
something because they were hesitant to vote "yes.")

An alternative is to multivalued votes per se is to have boolean votes but
to give someone the opportunity to specify how much they think their vote
should matter.  For example, they might say "yes" but that they're not an
expert and only wanted their vote to count, say, 30% instead of 100%.  Of
course this is isomorphic to actual multivalued votes if people get to
specify their own weight-percentages.

Anyone know of other theories of vote-tallying (more generally, quantified
opinions)?

--Todd

p.s.  Here's some interesting food for thought:  Came across this a while
back while looking up info on advanced hashing functions:

   http://ourworld.compuserve.com/homepages/bob_jenkins/vote.htm


[followups to lugnet.off-topic.geek]



1 Message in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR