[uf-discuss] hReview feedback

[Craig Ogg]

On 1/18/06, Paul Bryson <paul at msn.com> wrote:
> If 0 is the lower bound, then that is a 75% rating.  If 1 is the lower
> bound, then that is 67%, or a difference of -9%.
> 0...........100%
> 0..1..2..3..4
> 1...2...3...4

I think my math obscured my point:  a rating of 3 out of 4 is the same
to a *human* regardless of whether 0 or 1 is the lower bound.  Why do
people disagree on what the lower bound of a rating system is?  I
think Ryan says Amazon's lower bound is 1 because you can't enter a
lower rating (you have to click on a star to set the rating).  I also
believes that this is why he made it the default.  I think others
disagree because the user may have wanted to give it less than one
star but was constrained by the input mechanism -- IOW, the real lower
bound is zero but Amazon just doesn't allow that to be chosen any more
than they let 3.5 be chosen.

You later points on the subjectivity of ratings reinforce my belief
that giving ratings a lower bound give them a false mathematical
meaning that doesn't really exist. IOW, you cannot meaningfully
convert from a star rating to a % or other different # of stars rating
and think you are still representing the raters' intent.

That said, I had hoped my contribution would clarify things rather
than further muddying the waters.  I don't think I succeeded.  Back to
lurking...

Craig