24 
BULLETIN 1480, U. S. DEPARTMENT OE AGRICULTURE 
Table 8. — Analysis of prices received by producers of cotton in the State of 
Texas, December 1, 1925, by crop-reporting districts — Continued 
COEFFICIENT OF VARIABILITY, 
PER CENT 
Price, 
cents per 
pound 
District 
1 
District 
2 
District 
3 
District 
4 
District 
4a 
District 
5 
District 
6 
District 
8 
District 
9 
State 
14.0 
11.0 
11.1 
5.9 
5.3 
9.8 
11.8 
10.7 
8.0 
11.3 
PROBABLE ERROR OF THE AVERAGE PRICE, CENTS 
0.17 
0.14 
0.17 
0.2O 
0.09 
0.11 
0.20 
0.18 
0.21 
0.06 
RELATIVE PROBABLE ERROR OF THE AVERAGE, PER CENT 
1.03 
0.74 
0.89 
1.02 
0.46 
0.61 
1.10 
0.93 
1.10 
0.43 
(See Table 6 footnotes.) 
In the South Atlantic States Delaware headed the list, with reports 
from 100 per cent of the counties, and Florida came last with reports 
from about 30 per cent. In the South Central States the range was 
from 82 per cent of the counties in Tennessee to only 34 per cent of the 
parishes in Louisiana, whereas in the far Western States the varia- 
tion was from 80 per cent of the counties in Arizona to 24 per cent 
of those in Nevada. About two-thirds of all the counties in 29 
States were represented by one or more reports; 20 of these States 
were, in the North Central or Eastern States ; 6 were Southern States, 
and 3 were in the far West. In some of the Southern States, in 
Georgia, for example, counties are very small units. In many of 
the Western States comparatively few counties are agricultural. 
VARIABILITY AND SIZE OF SAMPLE 
State price averages apply to large areas and dissimilar conditions. 
It has been pointed out that farm prices tend to align themselves in 
zones somewhat similar to belts of rainfall. Freight rates, transpor- 
tation facilities, accessibility, lack of timely knowledge of market 
conditions and prices, surplus and deficit production, and differences 
in grade, quality, variety, age, and condition, are some of the factors 
that cause variation in the prices reported from different sections of 
a State. The farm prices of some products are much more variable 
than those of others. 
With some idea of the variability in a price sample, or the dis- 
persion of the individual price reports, it is possible to tell how many 
reports are necessary to obtain a given degree of reliability or 
stability in the average. The greater the variability, the greater 
number of reports needed to give stability to the average of the 
sample. 
On analysis it is found that the farm prices of wheat, corn, cotton 
hogs, butter, eggs, and wool, seldom show a coefficient of variability, 6 
6 The most common measure of the variability in a given sample is probably the 
standard deviation. It measures the range from the average within which approximately 
two-thirds of the reported prices will fall, assuming a normal or bell-shaped distribution. 
When the standard deviation is expressed as a percentage of the average, it is known as 
the " coefficient of variability." 
