L. ISSERLIS 
53 
4. Further we take : (7->.'«"'>" = 
We now collect into a single table the values of the reduced moments, the 
correlation coefficients and the correlation ratios. 
TABLE XIV. 
1x1^ 
•157282 
''a:;/ 
•907489 
Ixifl 
•343344 
•918157 
•019478 
IxV 
2-281729 
xVy 
•915430 
x'l y ' xy 
•014476 
^d>y 
2-253653 
2-537432 
ryz 
-830729 
q^^ 
•016995 
!/Vz 
•851196 
yTz - r-,y 
-034424 
2-531793 
zVy 
•833060 
zTii - >'^yz 
-003878 
■580509 
3-118877 
r.x 
•830463 
•027925 
x^z 
-844792 
xTz - r^zx 
•024005 
•030294 
-831447 
■001635 
b597889 
1-359207 
?xV 
1-605274 
•138335 
It will be observed that the differences ^rj^y — r^y^ and ^tj^^ — r^^^ are very small, 
showing that the regression lines of weight on age, and of height on age are nearly 
straight The six empirical regression lines are given in Figs. 1-6. 
Fig. 1. Regression of Age on Weight. 
Mean Weight. 
-I ■■■ I 1 1 I [ — ( I ■ - ( 1 J ■ ■ ■ 1 1 
r Mean Age 
1 . 1 1 1 
1 1 1 1 1 i 1 1 1 1 1 1 1 
26 31 36 41 46 51 56 61 66 71 76 81 86 9-1 96 101 106111 116 121 
Weight in lbs. 
