The overlapping coefficient                                     [sg27: STB-22]
---------------------------

        ^overlap^ var1 var2 [ ^if^ exp ] [ ^in^ range ]

        ^overlap^ variable [ ^if^ exp ] [ ^in^ range ]^ , by(^groupvar^)^

        ^overlapi^ mean1 mean2 sd

        ^overlapi^ mean1 mean2 sd1 sd2

        ^overlapi^ mean1 mean2 sd1 sd2 n1 n2

report the overlapping coefficient, a measure of the "overlap" of two distri-
butions.  Formally, the overlapping coefficient is a measure of the closeness
of the location of two distributions.  The overlapping coefficient is equal to
one minus the dissimilarity index.  The overlapping coefficient is equal to one
when the means of the two distributions are identical.  It approaches zero as
the means grow infinitely distant.

The first two syntaxes calculate the overlapping distribution for a variable
in the current data set.  The first syntax compares the location of the two
variables listed, while the second syntax compares the means of the variable
for the two levels of ^groupvar^.  The next three syntaxes pertain to the
immediate form of the command.  The first syntax reports the overlapping
coefficient under the assumption the two variables have identical variances.
The second syntax adjusts the overlapping coefficient for differences in the
variances of the variables being compared.  The third syntax reports both
forms of the overlapping coefficient.


Author
------
                Richard Goldstein, Qualitas, Inc.
                richgold@@netcom.com

Also see
--------

 Manual:  [5s] ttest
    STB:  sg27 (STB-22)
On-line:  ^help^ for ^ttest^