According to the comments in Dun's Review, the wide scatter of individual cases indicates a "growth tendency in favor of small concerns" rather than indicating that "all large stores had built sales volume more rapidly than the small ones."

It should be noted that the vertical rulings are logarithmic.

The limited number of vertical and horizontal rulings was intentional -that is, they were limited to make it easy for the reader to notice the lack of pattern of the dots.

CORRELATION CHARTS

. The Development of Electrical Advertising as Revealed by the Number of Watts Usea per Inhabitant in 143 Cities in the United States in 1922.

This scatter chart is supplemented by an average hne secured by computation.

The lack of pattern here indicates that there is little correlation between the two variables.

Aggregate population of the 143 cities was 6,300,000. The average of 3 1/3 watts per inhabitant was weighted according to population, not according to the number of cities.

■Hill

~i"

"

•

r

._llil«k

' •

•

. .

.

•1

" '

• .•

• '.

;<

.^^

L •

• ,

*

roruLATio,

1 1

« o» c

'

r

"

National Electric Light Association Bulletin, Feb. 1923.

CENTS

BUSHEL

'28

PRICES ADJUSTED FOR LEVEL OF INCOMES OF INDUSTRIAL WORKERS (l924-29= tOO)

300 320 340 360 380 400 420

PRODUCTION PLUS NET IMPORTS OR MINUS NET EXPORTS

(BUSHELS. MILLIONS)

U S Department of Agriculture. Bureau of Agricultural Economics. SCALE .9

B. The Relation of Adjusted Farm Price of Potatoes to Production in the United States from 1921 to 1936.