| |
In lugnet.off-topic.debate, David Eaton writes:
> In lugnet.off-topic.debate, Harvey Henkelman writes:
> > This should be interesting to watch...
> >
> > http://members.lugnet.com/polls/ballots/?n=121
>
> Tried to make a less "flawed" version of the poll:
> http://members.lugnet.com/polls/ballots/?n=122
>
> (How'd I do, Lar?)
IMHO, pretty good.
It may not be a *perfect* partition of the universe of possibilities into
equivalence classes, as you can argue that "yes, but..." always fits within
an unqualified "yes" and thus is a subset answer (the counter argument is
that the unqualified and qualified together imply partition). Also you
arguably left out "don't know" (arguably, because the counterargument is if
you don't know, don't answer!).
But it's vastly superior to Harvey's... as having a superset/subset answer
relationship (so that you have overlapping coverage but still cover the
universe) on answers that convey little or no bias on how the poll taker
felt is a LOT better than highly biased answers that don't even cover the
universe satisfactorily.
This avoids the questions of whether a poll was really necessary, what
prompted Harvey out of the blue to construct one, and so forth. But I'm just
answering what you asked.
(trimmed lugnet.general from the groups)
|
|
Message is in Reply To:
18 Messages in This Thread:
  
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
