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 line 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.
 | An image should appear at this position in the text. To use the entire page scan as a placeholder, edit this page and replace "{{missing image}}" with "{{raw image|Graphic methods for presenting facts (1914).djvu/217}}". Otherwise, if you are able to provide the image then please do so. For guidance, see Wikisource:Image guidelines and Help:Adding images. |  |
M. O. Lorens, in the Publications, the American Statistical Assn.
Fig. 161. Curves to Show the Percentages of the Total Population of Prussia in 1892 and in 1901 that Received Various Percentages of the Total Income as Considered on the Horizontal Scale
If incomes were all equal the relation of population and income would be
expressed by the straight diagonal line. The amount of inequality
between various incomes is shown by the amount the curve diverges
from the straight line. There was greater inequality of incomes in
Prussia in 1901 than in 1892
Imagine the whole population placed in a long line and ranked according to income. The people in this line could be counted off into several equal groups so that each group would contain say 10 per cent of the total number. It would then be simple to compute the income of each group as a percentage of the combined income for all groups. The resulting group percentages would be plotted cumulatively as the dependent variable on a chart for which percentages of population would be the independent variable. Fig. 161 unfortunately shows the independent variable used for the vertical scale. A better arrangement may be seen by observing the illustration through the back of the paper with the two zeros appearing at the lower left-hand corner.
