scipy.stats.tiecorrect

scipy.stats.tiecorrect(rankvals)

Tie correction factor for ties in the Mann-Whitney U and Kruskal-Wallis H tests.

Parameters :

rankvals : 1-d array_like

A 1-d sequence of ranks. Typically this will be the array returned by stats.rankdata.

Returns :

factor : float

Correction factor for U or H.

See also

rankdata

Assign ranks to the data

mannwhitney

Mann-Whitney rank test

kruskal

Kruskal-Wallis H test

References

[R200]

Siegel, S. (1956) Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill.

Examples

>>> tiecorrect([1, 2.5, 2.5, 4])
0.9
>>> ranks = rankdata([1, 3, 2, 4, 5, 7, 2, 8, 4])
>>> ranks
array([ 1. ,  4. ,  2.5,  5.5,  7. ,  8. ,  2.5,  9. ,  5.5])
>>> tiecorrect(ranks)
0.9833333333333333