Re: the midpoint

bob green (
Fri, 7 May 1999 22:00:14 +1000

Many thanks to all who responded to my original inquiry.

My interest at this point is largely from a research point of view, though
recently I have been looking at some grids compiled by colleagues and this
issue has come up on several occasions (both explicitly and also implicitly,
in that some constructs seemed to encompass more than two poles).

My hopefully, not overly simplistic reading of the replies is as follows:

careful completion of the grid, including the flexibility to split
constructs/elements, rename constructs and checking out the meaning of the
midpoint (if used) is necessary.

careful selection of analysis program/type of analysis, including the
facility to handle missing values/NA ratings and being aware of the impact
of inclusion of missing values etc on the type of analysis undertaken, is
also essential.

Thanks again, and any other comments are welcomed.


Bob Green