Hmm - well, as the one doing the choosing, I can assure you that in some instances, what you actually got was the result of 'rand([b,c,d])', which you'll have no option but to collate that as a vote for (e.g.) 'b', rather than a 1/3rd vote for b, c & d...
While I understand what you are trying for, from this side of the survey, it /seems/ that it'll give you flawed results! Hopefully there'll be enough responses to mean that it's not so, though. /Gwyn On Monday, July 9, 2007, 11:53:02 PM, Clinton <[EMAIL PROTECTED]> wrote: > No information is lost. It's completely dependent upon what > information we're looking for. And what we're looking for is the MOST > important things to people, not the top three, or even the top two. > Just one and only one. > I want to force you to choose. > Worry not, nothing will be left out. > Cheers, > Clinton > On 7/9/07, Gwyn Evans <[EMAIL PROTECTED]> wrote: >> Hmm, not sure I really agree with you there - having done the survey, >> there were a number of questions where having to choose just 1 item >> felt as misleading as choosing all would have been. >> >> The correct answers would have been a few of the items. You may >> believe that one always stands out, but certainly with my answers, >> there are answers where you will see an answer as 'b', for example, >> but in reality, it could have been 'b','c' or 'd' but not 'a' or 'e'. >> There's information being lost there!
