30 
BULLETIN 1480, U. S. DEPARTMENT OF AGEICITLTUKE 
crop is normally sold as grain, and sales of seed are only a very small 
part of the total sales; but with oats, the sales for seed are relatively 
more numerous, and hence appear more frequently in a price sample. 
In some localities the sales of seed oats may be the only sales. 
Table 11. — Farm prices of oats: Selected illustrations of size of sample, meas- 
ures of dispersion, and probable error 
[Per bushel] 
State and date 
Average 
Number 
of reports 
price 
(arith- 
metic) 
mean) 
Standard 
deviation 
of reports 
Cents 
Cents 
SI 
49 
5.7 
42 
46 
0. 
27 
51 
6.0 
101 
47 
6.9 
53 
47 
5.0 
24 
57 
8.5 
59 
26 
4.0 
53 
25 
1.5 
29 
88 
16.5 
247 
37 
3.4 
66 
49 
7.0 
Coeffi- 
cient of 
varia- 
bility 
Probable 
error of 
Relative 
the aver- 
probable 
age price 
error 
or mean 
Cents 
Per cent 
0.4 
0.9 
.6 
1.2 
.8 
1.5 
.5 
.9 
. 
1.0 
1.2 
2.1 
.4 
1.4 
2 
. 7 
2.1 
2.4 
.1 
.4 
.6 
1.2 
Four 
times 
relative 
probable 
error l 
Indiana, June, 1925 
Illinois, June, 1925 
Missouri, June, 1925 
New York, October, 1925 
Pennsylvania, November, 1925. 
Maryland, March, 1926 
North Dakota, October, 1925 
(price range, 20 to 40 cents) . .. 
North Dakota, October, 1925 
(price range, 20 to 29 cents) _.. 
Georgia, October, 1925 
North Dakota, December, 1925 °< 
Maryland, December, 1925 2 ... 
Per cent 
11.6 
11.9 
11.7 
14.8 
10.6 
15.0 
15.3 
6.1 
18.7 
9.1 
14.3 
Per cent 
3.0 
4.8 
6.0 
3.6 
4.0 
8.4 
5.6 
2.8 
9.6 
1.6 
4.8 
1 The probabilities are ninety-nine out of one hundred that the average of a much larger sample collected 
in the same way and at the same time would not vary from this average by more than four times the prob- 
able error. 
2 The Dee. 1 prices were reported by crop reporters, and not by the regular price reporters, who report 
on the 15th of each month. 
The larger size of the sample for the December 1 prices renders 
the averages for that date more stable than is the averages of the 15th- 
of-the-month prices. Four times the probable error is less than 5 
or 6 per cent of the monthly price averages for the larger States, 
including New York, and Pennsylvania, which are not thought of as 
surplus oat States. For both Maryland and Georgia, the sample was 
not sufficiently large to hold the probable error down to a point where 
four times the relative probable error is less than 6 per cent, but 
even in those States, four times the relative probable error does not 
exceed 10, per cent. On the whole, oat prices are reasonably reliable. 
COTTON AND COTTONSEED 
In looking over cotton prices for the several Southern States 
(Table 12), one is impressed with the smallness of the coefficients 
of variability of the prices of cotton lint, which is sold on a highly 
organized local market, as compared with the variability of the 
prices of cottonseed, which the farmer considers a by-product. The 
number of reports on cotton prices is relatively small for most of the 
States, but the variability is sufficiently low to keep the probable 
error at a point where four times the probable error ranges from as 
low as 1.2 per cent to not more than 7 or 8 per cent. Cottonseed 
prices seem to share the great variability of most farm prices for 
Southern States other than those of cotton lint. As the season 
advanced from August, with few sales, to October, when the season 
is at its height, the number of reports received increased materially. 
Cotton lint is the only important crop none of which is consumed 
on the farm or used for seed. The supply of American cotton is 
