-------------------------------------------------------------------------------
help for domdiag
-------------------------------------------------------------------------------

Dominance diagram

        domdiag varname [if exp] [in range] , by(byvar) [ reverse
                 graph_options ]


Description

    domdiag plots a dominance diagram given a division of the values of the
    numeric variable varname into precisely two groups by the categories of
    byvar.

    Suppose there are n_1 values in the group for the lower value of varname
    (hereafter called group 1) and n_2 values in the group for the higher
    value of varname (group 2). Then there are n_1 * n_2 pairwise
    comparisons, which are plotted on a graph with axes rank on varname
    within group 1 and rank on varname within group 2. The symbols used are
    "+" if value in group 1 > value in group 2, "0" if value in group 1 =
    value in group 2, and "-" if value in group 1 < value in group 2. This
    helps to make graphic what is done in ranksum: the proportion of pairs
    symbolised "+" plus 1/2 the proportion of pairs symbolised "0" is the
    estimate that P((varname | group 1) > (varname | group 2)).

    By default, ranks for the group with fewer values are plotted on the y
    axis.


Remarks

    This graph was used by Cliff (1993, 1996). Some earlier related ideas are
    reviewed by Fisher (1983). Newson (2002) gives a broader discussion and
    an example implementing domdiag.

    domdiag temporarily makes a dataset of n_1 * n_2 observations.  This may
    be very much larger than the existing dataset and the user should make
    sure that enough memory is allocated.


Options

    by(byvar) specifies subdivision into two groups and is a required option.
        If byvar defines three or more categories and it is desired to focus
        on just two, make sure the others are excluded by using if.

    reverse specifies that axes are to be reversed from the default.

    graph_options are options of scatter.  By default, note() is used to
        display P((varname | group 1) > (varname | group 2)).


Examples

    . domdiag mpg, by(foreign)


Author

    Nicholas J. Cox, Durham University, U.K.
    n.j.cox@durham.ac.uk


References 

    Cliff, N. 1993. Dominance statistics: ordinal analyses to answer ordinal
        questions. Psychological Bulletin 114, 494-509.

    Cliff, N. 1996. Ordinal methods for behavioral data analysis.  Mahwah,
        NJ: Lawrence Erlbaum.

    Fisher, N.I. 1983. Graphical methods in nonparametric statistics:  a
        review and annotated bibliography. International Statistical Review
        51, 25-38.

    Newson, R. 2002. Parameters behind "nonparametric" statistics:  Kendall's
        tau, Somers' D and median differences.  Stata Journal 2: 45-64.
        http://www.stata-journal.com/sjpdf.html?articlenum=st0007



Also see

 Manual:  [R] signrank
On-line:  signrank