Point Sampling Surveys for Potato Acreage in Colorado's San Luis Valley

Abstract

DURING THE PAST FEW years considerable attention has been given to the collection of agricultural data by probability sampling. Enumerative surveys employing a large group of trained interviewers often are used to obtain such data. Multiple frame surveys using lists of specialized producers to supplement an area frame may be employed, or several lists may be used to screen the population for the elements with the desired characteristics. In some situations, less sophisticated probability techniques requiring a relatively small group of trained people may offer possibilities of rapid and efficient execution in the field. If there is a problem of available resources survey work, the technique of point samng may provide an alternative which should be considered. Point sampling is a method of observing the use of very small units of areas of equal size whose locations are determined by random points. The sizes of the fields in which the random points fall are immaterial. This paper gives an example of the application of the technique with the related costs and variances. It presents a brief summary of the survey methodology used during the first 2 years of a project in Colorado. The project was initiated in June 1964 in response to industry desires for an early season forecast of potato acreage for the San Luis Valley in Colorado. The statistician in charge of the Colorado office of the Statistical Reporting Service, R. S. Overton, and his staff directed the survey in the field and assembled source materials for the frame. The county offices of the Agricultural Stabilization and Conservation Service were the source for aerial photographic coverage of the sample units selected for the survey. The Sampling Frame