*********************************************************************************
THIS PROGRAM WAS CREATED FOR FREE DISTRIBTUION, AND ALL THAT USUAL STUFF BY:

Malak........

There are notes at the foot of the page
*********************************************************************************
*********************************START OF CODE**********************************

10:CLRHOME
20:DIM L1->N
30:1 ->A
40:FOR(T,1,N,1)
50:T->L6(T)
60:END
70:SORTA(L1,L2,L6)
80:FOR(A,1,N,1)
90:A->L3(A)
100:END
110:1->B
120:SORTA(L2,L1,L3,L6)
130:FOR(B,1,N,1)
140:B->L4(B)
150:END
160:SORTA(L6,L1,L2,L3,L4)
170:1->C
180:(L5(C))2->L6(C)
190:L6(C)+D->D
200:END
210:DISP "N=",N
220:DISP "D2=",D
230:(1-((6*D)/(N(N2-1))))->R
240:DISP "R=",R


*********************************END OF CODE************************************

Notes:

1: The numbers before each line are NOT for the calculator, but are intended for ease of viewing and referencing.
2: On lines 180 and 230, there is a squared symbol, and not a 2.
e.g. line 180 should read (L5(C)) squared.
      line 230 should read (1-((6*D)/(N(N squared-1))))->R

3: This program uses the lists to find the correlational coefficient.

These are to be set before starting the program
L1 is the first variable.
L2 is the second variable.

After running the program:

L1 is the first variable.
L2 is the second variable.
L3 is the rank position of variable one.(L1)
L4 is the rank position of variable two.(L2)
L5 displays the difference between ranks.
L6 displays the difference between ranks squared.

On the screen, after running the program, the following will be displayed:

n = the number of items (sets of data).
Ed = Sum of the differecne between value's squared.
r = The Spearmans' correlational coefficient.

*********************************************************************************

Problems

1: Due to the limits of this calculator, the TI-80, it takes a long time to calculate the Spearmans' correlational coefficient.  There also maybe a problem if there are a lot of fields (maximum tested is about 40).
2: Tied ranks will distort result.  If you know about this test, then you will realise this.  Therefore, you will need to make sure there are no ranks that are the same.
(Someone could add a tied rank checking routine, but this would slow the program down)

*********************************************************************************

If you like this program, and want to contact me:

e-mail 		milkandbiscuit@icqmail.com
icq		98474936
