NSKAPPA MACRO

The nskappa.sps macro computes Cohen's Kappa for a two way table, which
does not have to be square (i.e., although conceptually the table would
be square, empirically, the raters may not have both used all of the
available categories, resulting in a table with empty rows and/or
columns). Statistics produced are estimates of Kappa, the asymptotic
standard error under the null hypothesis of a 0 population value (ASE0),
a Z-value and approximate significance level (p-value) for testing the
null hypothesis, the asymptotic standard error under the alternative
hypothesis (ASE1), and 95% confidence interval bounds using ASE1.

The nskappa.sps macro is most easily used by simply having it resident as 
a text file in your working directory and executing the following SPSS 
syntax:

INCLUDE NSKAPPA.SPS.
NSKAPPA VARS=rowvar colvar.

where rowvar and colvar are the two variables containing the ratings.
Note that the two variables must be numeric, and they must be coded as 
consecutive integers from 1 to the number of rating categories (though
neither variable must have all categories represented). Data can be set
up as case by case data or as aggregated or summary data with one case
per cell of the table and a WEIGHT variable in effect. (Weights are not
rounded in the macro, so the Kappa value produced here will differ from
that given by CROSSTABS in Releases 6.1.2 and above if the weights for
a cell do not sum to an integer.)

The macro first saves your working data file to a file named ka__tmp1.sav.
The double underscore in the file name is an attempt to render unlikely the
overwriting of an existing file. The SET commands are used to minimize
output; they may be changed or removed if you have problems running the
macro in order to aid in identification of problem sources.actions

The macro should function on any SPSS release offering macro support and the
MATRIX procedure (which means for modular systems you must have the Advanced
Statistics module). It creates two new files, ka__tmp1.sav and ka__tmp2.sav,
which contain respectively your original data, and a file used to store the
values printed in the reported output.

Multiple invocations of the macro in the same interactive session will
produce the following note:

>The macro name specified on the DEFINE command duplicates the name of a
>previously defined macro.  This instance will take precedence.

This warning does not indicate a problem and may be ignored.