16 
BULLETIN 1277, U. S. DEPAETMENT OF AGRICULTURE 
own feet." Such analysis as here outlined requires both more 
cases and more precise data than have usually been obtained. Prob- 
ably the route method, or a combination of route and survey 
methods, will be necessary to get the full information needed. 
In the meantime, results from engineering studies of farm machinery 
can be used to supplement the farm-survey studies. 
The use of average hours of labor is but a very vague way of 
measuring labor input. For both analysis and practical application, 
the measurement of intensity in the use of labor must be brought 
down to a solid, definite basis of operations performed, intensity of 
each operation, and specific causes of variations in labor requirement 
of each operation. 
Xo doubt a considerable part of the variation is also due to 
the errors of estimate sure to accompany the survey method; and 
far from being eliminated by route methods. There is always a 
considerable amount of " idle time " which will be charged to 
different tasks on different farms. 
OUTPUT PER UNIT OF INPUT 
The analysis thus far has shown the variation in inputs, and has 
indicated the ways in which the inputs of the different factors are 
correlated. The next step is to discover what output accompanies a 
given input of each of the elements of input. This is really the 
fundamental problem in the analysis of input — our knowledge of the 
relative profitableness of different practices can be no more accurate 
than our knowledge of the effect of those practices upon the product. 
POTATO PKODUCTIOX ANALYSIS 
It might seem at first that all that is necessary for this new purpose 
is to add two columns to the frequency tables presented in the last 
section, making them appear as shown in Table 16. 
Table 16. — Apparent effect of fertilizer on potato yield 
Amounts of input— value of fertilizer 
Number 
of farms 
Output- 
yield per 
acre 
Output 
per 
dollar's 
worth of 
fertilizer 
$0.00to$5.49 . ... 
44 
121 
76 
43 
Bush f Is 
103.1 
114.1 
132. 9 
132. 1 
Bushels 
26.4 
-S5.50toS10.49 .. 
15.0 
S10.50toS15.49 
10.4 
S15.50 to S20.49 
7.5 
Unfortunately, however, it may be that the farmers using the 
large amounts of fertilizer are using more or less intensive cultural 
methods, more labor and less equipment, or vice versa: and the 
differences in yield may be due to any one or all of these several 
differences in input. By proper statistical methods (discussed in 
detail on pp. 40 to 44) it is possible to eliminate the effects of the 
other inputs, and to determine the net effeel of changes in the input 
of fertilizer alone. Tables 17. IS. and 19 show some of the results 
obtained from the potato data, as far as it was possible to isolate the 
factors in the data at hand. These relations are shown graphically 
in Figure U. 
