Google
 

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