Thanks as usual (but not less serious) to Kit Baum, a new version of
-scores- is available from SSC.
-scores- generates a new variable of scores (row-wise) of a set of
variables allowing the specification of the number of valid values
required by using row functions of -egen-.
The new version also allows to calculate median and percentile scores,
proportions of maximum possible scores, and shrunken proportions of
maximum possible scores - the latter are useful for dependent variables
in beta regression models, see Smithson & Verkuilen (2006, p. 54f.) and
-findit betafit-.
Additionally, -scores- now allows by: and fweights, aweights, or
iweights (only useful if requesting z-scores, mean centered scores, or
shrunken proportions of maximum possible scores).
Reference:
Smithson, M. & Verkuilen, J. (2006). A better lemon squeezer?
Maximum-likelihood regression with beta-distributed dependent variables.
Psychological Methods, 11, 54-71.
