On Wed, Feb 27, 2013 at 5:49 AM, Nick Cox <njcoxstata@gmail.com> wrote:
> No one seems interested in this suggestion.
To the contrary, I was thinking a lot about the problem but didn't
have anything coherent to say.
But it is well to know
> what patterns do or do not exist before you start quantifying them.
It's essential!
> This is just to underline that repeating the graphs with ranks
> underlines how much information is thereby discarded and that while
> such graphs may not work well with many more raters and/or scores,
> they do work well here.
The trick with many more raters or scores isn't to abandon graphing,
but to choose a good projection to highlight features of interest.
Many areas, such as agreement modeling or reliability measurement, are
not viewed as model fitting or building enterprises. This needs to
change.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
