[EM] RE : Naive question about Range Voting -- why 0-99 and not0-100?

[Bob Richard]

Another reason for this is based on experience in public opinion
research.  Scales with an exact midpoint (0 to 10) encourage people to
pick the midpoint to avoid committing themselves.  Scales with an even
number of choices (0 to 9, or 1 to 10) make people come down on one side
or the other, even if only slightly.

Back when I was involved in questionnaire design, researchers had pretty
much settled on the format, "On a scale from 1 to 10, how would you rate
X?"

--Bob Richard

> -----Original Message-----
> From: election-methods-bounces at electorama.com
> [mailto:election-methods-bounces at electorama.com]On Behalf Of Kevin
> Venzke
> Sent: Monday, November 20, 2006 12:19 PM
> To: election-methods at electorama.com
> Subject: [EM] RE : Naive question about Range Voting -- why 0-99 and
> not0-100?
>
>
> Monkey,
>
> --- Monkey Puzzle <araucaria.araucana at gmail.com> a écrit :
> > Hi folks,
> >
> > Is there any reason to not allow range voting with scores between 0
> > and 100?  If the intent is to allow a full range of voter
> expression,
> > why not go the extra point?  I thought the whole idea was to avoid
> > imposing a structural restriction on the voters.
>
> Warren came to the conclusion after a survey he conducted that 0 to
> 100 gives some the impression that the assigned scores must total 100.
> He supposes that a ceiling of 99 would lessen this impression.
>
> See http://www.math.temple.edu/~wds/homepage/works.html, paper 82.
>
> Kevin Venzke