BUREAU OF CROP ESTIMATES. 421 



with local con<liti(>iis and they hu\e no basis for making crop reports 

 which will be (;omj)arablc with preceding years;' (3) their period of 

 service in some counties is relatively short, they are constantly 

 shifted about and tlie turnover is frequent, wdiich prevents them 

 from being in position to furnish comparable data one year with 

 another; (4) statistical training and experience is not one of the 

 (]ualilications required for appointment as county agents, yet statis- 

 tical work is as much a specialty as accounting, or horticulture, or 

 commercial fruit growing; (5) while some county agents have a 

 natural inclination to take an interest in the statistical side of pro- 

 duction and therefore can be used to splendid advantage by the 

 bureau, they are few in number and it is exceedingly difficult to 

 interest the relatively large number of county agents who have no 

 such inclination; (6) county agents are employed primarily for 

 entirely dilfcrent purposes than collecting and summarizing statisti- 

 cal data, their main pur})ose being to promote better and more 

 profitable methods of farming, which is a job of sufficient size to 

 absorb all their energies: (7) in many counties practically their entire 

 salaries are paid from the local funds and their responsibility is 

 directly to local authorities. Because county agents ni-e fully occu- 

 pied if not overburdened with duties which have little or no relation 

 to crop estimating, exj^erience has shown that on the average the num- 

 ber of replies received from inquiries addre&sed to county agents is 

 less than from other classes of repoi'ters. The reluctance of county 

 agents to report to the bureau in some cases and their inability to 

 report on account of other duties in many cases, and the fact that 

 nearly one-third of the counties are without any coimty agents, are 

 fatal to any plan of relying upon them as a princi})al or main 

 source of data for crop estimating on a uniform and country-wide 

 basis. In the circumstances they can be utilized to advantage only as 

 ;i, supplemental source of information. 



Any plan to utilize county agents successfully in crop-estimating 

 work, with exceptions in a few States only, where conditions are 

 especially favorable, will involve increased expense for necessary 

 supervision, instruction, training, administration, and for summa- 

 rizing and interpreting data supplied by them. The situation with 

 respect to utilizing county agents in crop-estimating work has been 

 thoroughly canvassed with the States Relations Service and the limi- 

 tations mentioned arc recognized. Several plans are under consid- 

 eration for development, with a view to the wider utilization of croj) 

 reports by farmers through encouragement and assistance by the 

 county agents, whenever the necessary funds are provided without 

 which the plans can not be made effective. One of these plans in- 

 volves the employment of at least four regional extension s|)ecialists 

 in crop estimating whose entire lime would l)e devoted to instructing 

 county agents and extension leaders in the technique of crop esti- 

 mating and educating them in the value, interpretation, and prac- 

 ticjil use of crop'repoi'ls in the l»usiness of fai'ming. 



ACCURACY OF THE ( ROI' ESTIMATES. 



The ^alne of the crop and live-stock estimates depends upon their 

 accuracy, timeliness, and amount of detail. The degree of accuracy 



of the estimat(\s can b(> (leterniinecl only ajiproxiniately for most 



