The purpose of this study was to determine a sampling procedure by which agricultural information may be efficiently obtained for areas smaller than the state, particularly the county. This study, therefore, is one of a series (2) (4) directed toward the problems of methodology of an agricultural sample-census. The objectives (8) of such a sample-census are: To obtain useful information (a) more frequently (annually instead of quinquennially, as at present), (b) of greater extent (include items not covered by the regular census) and (c) at greater economy. Consequently the sample-census is proposed at least in part to supplement the regular census in providing the information needed in the analysis and understanding of special agricultural problems arising in the field of policy and programs, such as farm tenancy, soil conservation, etc. This study, however, is limited to an investigation of merely a few of the many problems of sample-census procedure. Here we are interested primarily in the following questions:
1. Is it possible to find a township or group of townships within a county which will satisfactorily represent that county for a period of years?
2. For a given degree of accuracy, how large must a sample he for (a) a county, (b) a crop-reporting district (a group of about 10 counties)?
3. Are there any satisfactory criteria by which a county may be stratified and thereby increase sampling efficiency?
This study provides, we believe, some aid in answering these and allied questions.
Strand, Norman V. and Jessen, Raymond J.
"Some investigations on the suitability of the township as a unit for sampling Iowa agriculture,"
Research Bulletin (Iowa Agriculture and Home Economics Experiment Station): Vol. 26
, Article 1.
Available at: https://lib.dr.iastate.edu/researchbulletin/vol26/iss315/1