.-
help for ^grvar^                                               (STB-42: gr28)
.-

A graphical procedure to test equality of variances
---------------------------------------------------

	^grvar^ sd n [^if^ exp] [^in^ range] [^,^ ^id(^strvar^)^ ^level(^#^)^ graph_options ]


Description
-----------

^grvar^ provides a graphical display for testing equality of variances
of normal populations. This method has been developed by Rao and 
Krishna (1997) following Ott's (1967) Analysis of Means procedure. 
This graph tests equality of several variances of normal populations, 
simultanously demonstrating the statistical and engineering significance. 

^sd^ is the standard deviation from the individual sample or study, 
and ^n^ is its sample size. 

If you have a dataset which contains data for all studies, then the @byvar@
command can be used to derive the standard deviations and sample sizes for the
individual samples. For example:

   . ^sort sample^
   . ^collapse (count) n=varname (sd) sd=varname, by(sample)^ 
   . ^grvar sd n, id(study)^

Alternatively, the @generate@ command may be useful.


Options
-------

^id(^labelvar^)^ supplied any variable, which is used to label the studies. 
If the data contains a labeled numeric variable, it can also be used.

^level(^#^)^ specifies the level of significance of the test, in percent.
The default is ^level(5)^.

graph_options are any of the options allowed with ^graph^, ^twoway^ except 
^xlabel()^, ^t2()^, ^b2()^, ^yline()^, ^symbol()^, ^pen()^, and ^connect()^.


Author
-------

       Aurelio Tobias
       Institut Municipal d'Investigacio Medica
       Barcelona, Spain
       email: atobias@@imim.es


References
----------

Rao, G. V. and S. H. Krishna. 1997. A graphical method for testing the equality
of variances. Journal of Applied Statistics 24: 279-87.

Ott, E. R. 1967. Analysis of means - a graphical procedure. Industrial Quality
Control 24: 101-109.


Also see
--------

    STB:  STB-42 gr28
On-line:  help for @graph@, @byvar@, @collapse@, @oneway@