Graphic Methods for Presenting Facts
If there is not the same number of orders in all groups or zones the curves will take some other shape and the shape will depend on the peculiar distribution of orders as may be seen in Fig. 157, Fig. 158 and Fig. 159
FREQUENCY CURVES
the shape of curve "A" and curve "B" in Fig. 159. By comparing Fig. 160, we can see that if the distribution of orders in Fig. 159 had been uniform, curve "A" would have been a straight diagonal hue and curve "B" would have been a curved line bowed upward instead of bowed downward.
Though Fig. 161 somewhat resembles Fig. 160, it is nevertheless constructed on an entirely different plan. In Figs. 157, 158, 159 and 160 the independent variable related only to size of order. For charts of the type shown in Fig. 160 the independent variable is a percentage. The dependent variable is also expressed as a percentage.
Imagine the whole
] I* ■r»lfir»Arl in M. O. Lorenz, in Ihe Publications, the American SlalUHcal Assn.
^/ ,. ' ^ ^ , Fig. i6i. Curves to Show the Percentages of the Total
Population of Prussia in 1892 and in 190 1 that Received Various Percentages of the Total Income as Considered on the Horizontal Scale
L. <u
E60
<to 50
0)
O40
/
/
/
/
•
•7'
f
/
/
/
/
:/ .•/
/
/
^
30 40 50 60 70
Percents of Total I ncome
80 90 100
EQUALITY PRUSSIA 1892. PRUSSIA 1901