Sorry, an error occurred while loading the content.

## Re: [RangeVoting] VotePoll: A few comments on the polls

Expand Messages
• ... I d like to nominate another, because I like fine-grained RV, and I think it s clunky to have multiple polls for one candidate. 6. RV101(kludge) Use an
Message 1 of 4 , Sep 30, 2005
• 0 Attachment
On 9/30/05, MIKE OSSIPOFF <nkklrp@...> wrote:

> That makes five ways of approximating or implementing 0 to 100 RV, given the
> current circumstances:
>
> 1. RV25 (0, 4, 8, 12, ...96)
> 2. RV26 (Default 0 and 4, 8, 12, ...100)
> 3. RV100 (Four 25-slot ranges)
> 4. RV101 (Default 0 and four 25-slot ranges)
> 5. RV101(Voted at the mailing list, to conveniently implement it)

I'd like to nominate another, because I like fine-grained RV, and I
think it's clunky to have multiple polls for one candidate.

6. RV101(kludge)
Use an approval ballot with the following choices:
0,1,2,3,4,5,6,7,8,9,10,20,30,40,50,60,70,80,90,100
Voters are requested to be nice to the poll taker and not click more
selections than necessary to express their vote. If the selected
numbers add up to >100, the rating is adjusted to count as 100.

Cheers,
- Jan

P.S. There are lots of other schemes that could be proposed, but I'm
not nominating them:

* Two plurality polls for each candidate, with choices 0,10,20...100
for the first poll and 0,1,...9 for the second. If you don't want to
check for votes >100, eliminate the 100 and make it RV100.

* One plurality poll per candidate, with choices
0,1,2,3,4,5,10,20,30,40,50,60,70,80,90,95,96,97,98,99,100
This would accomodate voters who only want fine-grainedness at the
ends of the range.
• ... There is no need, with a relatively sophisticated population of voters, to have so many choices. If you must have granularity 99, why not the following
Message 2 of 4 , Oct 1, 2005
• 0 Attachment
At 01:24 AM 10/1/2005, Jan Kok wrote:
>6. RV101(kludge)
>Use an approval ballot with the following choices:
>0,1,2,3,4,5,6,7,8,9,10,20,30,40,50,60,70,80,90,100
>Voters are requested to be nice to the poll taker and not click more
>selections than necessary to express their vote. If the selected
>numbers add up to >100, the rating is adjusted to count as 100.

There is no need, with a relatively sophisticated population of
voters, to have so many choices. If you must have granularity 99, why
not the following choices:
64,32,16,8,4,2,1.

As to the poll taker, summing the votes is no more difficult with
binary than as many decimal slots, and it is easier than decimal for
the same granularity. In Range, the individual ballots are not
necessarily counted, just the sums. There is no need to compile
individual scores.

Frankly, I'd reserve 64 for an approval cutoff, being considered the
minimum Approval score. Scores below that do indicate a kind of
preference, but a preference in rejection.

As to how to vote for more than one candidate on a ballot, one can
present multiple candidates as options.

Candidate A, 64:
Candidate A, 32:
...
Candidate A, 2:
Candidate A, 1:
Candidate B, 64:
Candidate B, 32:
etc.

Yahoogroups allows 25 options in a single poll.

I recommend experimenting with coarser granularity. A 5-option poll
can thus be run Range with granularity 32, which is really
vanishingly close to fine-grained. If you must have granularity >=
100 (why?), then you need 7 slots and you can have three options in a
single poll.

Another method would be to use two poll questions, one for positive
ratings and one for negative. Effectively, the 64-bit would be
expressed by which poll you answered on. It's pretty easy, in a
browser, to have more than one poll open, so you could really do both
at the same time.
• Dear Mike Ossipoff, ... If you want to exclude all those people from the polls who have somewhen somewhere cast a preference for some other election method,
Message 3 of 4 , Oct 1, 2005
• 0 Attachment
Dear Mike Ossipoff,

you wrote (1 Oct 2005):

> Markus Schulze voted for BeatpathWinner in the current
> poll. As much as I'd like to count Markus's vote,
> because he voted for something that I like, I don't
> know if Markus advocates RV. If he doesn't, then
> I don't know if I have a right to count Markus's
> BeatpathWinner vote in a poll on how to vote in
> an RV advocacy organization.

If you want to exclude all those people from the polls
who have somewhen somewhere cast a preference for some
other election method, then, of course, you have to
exclude almost everybody.

Markus Schulze
Your message has been successfully submitted and would be delivered to recipients shortly.