A. RlTCHIE-8C0TT 
•524 
H = 
•482 ± 
•0631 
A.P.E. 
= 
•5018 ± 
•0235 \ 
/■„ - -501 ± 
•030 
(•030) 
I'c = 
•5173 ± 
•2330 
/i2 = -499 ± 
•028 
(•029) 
•5030 ± 
•0239 
■ = -508 d: 
•035 
(•032) 
>w = 
•5025 ± 
•0236 , 
=^ -504 ± 
•031 
(•032) 
^0 = 
■5098 
Vh2 
Mean 
dispersion 
•5211 
•4885 
■4956 
•5115 
Individual 
dispersion 
■4798 
■4907 
■4709 
•4553 
E. Table I divided so that the frequency of differs very little from the 
frequency of a table with the same marginal frequencies but of zero correlation. 
The mean is iij cell n.^^. 
1+2 
3 
4+5+6+7+8 
Total 
1 
27 
5 
2 
34 
2 
166 
79 
56 
301 
3+4+5+6+7+8 
165 
180 
320 
665 
Total 
358 
264 
378 
1000 
Here 
301 X 264 
1000~ 
= 79^464 so that the constant term in the equation for / is a 
small quantity and any error of sampling will have an excessive weight. It will be 
found as one might expect that the p.e. of is very large. 
The very large value of Ty^^ is due to the column having the marginal total 378, 
for the frequency 2 in it is the nearest whole number to a true value and being 
so small, a small absolute difference makes a large fractional value resulting in 
a large difference between the true and apparent standard deviations of this par- 
ticular array. Actually the method applied is inapplicable to a frequency of this 
order. 
r^ = 
•502 
H = 
•4827 ± 
•0246 
A.P.E. 
^■p = 
•4991 ± 
•0245 ^ 
r„ = ^500 ± 
•056 
(•054) 
= 
•4658 ± 
•4371 
/■i2 = -498 ± 
•029 
(•029) 
= 
•4995 d= 
•0327 
■ r^i = -500 ± 
•072 
(•054) 
'«) 
•4995 ± 
•0247 
•r^a = -500 ± 
•030 
(•029) 
U = 
•5065 
Mean 
dispersion 
•4862 
•5169 
•4950 
•4966 
Individual 
dispersion 
•4906 
•4991 
■5022 
•7150 
Biometrika xii 
