84 
PROFESSOR K. PEARSON AND DR. A. LEE ON 
I now proceed to determine the correlation in the errors made in determining the 
ranges corresponding to any two classes of any two variables which are correlated. 
For this purpose let the frequency correlation table be dressed as follows, in the 
diagram below. 
Here m-,j denotes the frequency of individuals common to the two classes and 
n j. Let My denote its “ conjugate,” or all the frequency which appears in neither n L 
nor n'j ; then 
As before, we have 
V 2 
V 2 
N = My + n, + n'j — m ;j 
Ui( N — ni) 
N 
My (N - My) 
N 
y 2 
m.j ( N - m;j) 
N 
(xvii.). 
(xviii.). 
(xix.). 
Further, since nq,- and My are mutually exclusive, we have 
v v p _ _ —''-A 
^My^ny^Myrny yr 
(XX. j. 
From (xvii.) we have for small variations 
S11; + Bn'j = S//iy — 8 My. 
Hence 
9V V P _ v 2 I V 2 _ V 2 _ V 2 _ 9^ V P 
^ — —'«'y U —'My —-'fly 'my—My i - t 
Myi*;; 
(xxi). 
