Trailing-Edge
-
PDP-10 Archives
-
decus_20tap2_198111
-
decus/20-0026/utest.doc
There are 2 other files named utest.doc in the archive. Click here to see a list.
SUBROUTINE UTEST
PURPOSE
TEST WHETHER TWO INDEPENDENT GROUPS ARE FROM THE SAME
POPULATION BY MEANS OF MANN-WHITNEY U-TEST
USAGE
CALL UTEST(A,R,N1,N2,U,Z,IER)
DESCRIPTION OF PARAMETERS
A - INPUT VECTOR OF CASES CONSISTING OF TWO INDEPENDENT
GROUPS . SMALLER GROUP PRECEDES LARGER GROUP. LENGTH
IS N1+N2.
R - OUTPUT VECTOR OF RANKS. SMALLEST VALUE IS RANKED 1,
LARGEST IS RANKED N. TIES ARE ASSIGNED AVERAGE OF TIED
RANKS. LENGTH IS N1+N2.
N1 - NUMBER OF CASES IN SMALLER GROUP
N2 - NUMBER OF CASES IN LARGER GROUP
U - STATISTIC USED TO TEST HOMOGENEITY OF THE TWO
GROUPS (OUTPUT)
Z - MEASURE OF SIGNIFICANCE OF U IN TERMS OF NORMAL
DISTRIBUTION (OUTPUT)
IER- 0, IF NO ERROR.
- 1, IF ALL VALUES OF ONE GROUP ARE TIED.
REMARKS
Z IS SET TO ZERO IF N2 IS LESS THAN 20
SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED
RANK
TIE
METHOD
DESCRIBED IN S. SIEGEL, 'NONPARAMETRIC STATISTICS FOR THE
BEHAVIORAL SCIENCES', MCGRAW-HILL, NEW YORK, 1956,
CHAPTER 6