ON APPLYING AREA AND MULTIPLE FRAME SAMPLING METHODS IN A WIDE RANGE OF BASELINE AGRICULTURAL AND RURAL SURVEY PROGRAMMES

In this paper we discuss the important role that complex, probability survey designs, and particularly area and multiple frame sampling methods, can play in ensuring accurate data and appropriate agriculture and rural development policies. We refer to large-scale multiple-purpose surveys in a wide variety of statistical programmes in developing countries. We think that these very important survey methods have not been taught, applied or disseminated properly. This paper recommends specific changes to improve the teaching of survey sampling. It also proposes to apply a concrete survey sampling approach, whenever possible: to construct a master area sampling frame on which to base and conduct the area frame survey designs of some of the baseline agriculture and rural survey programmes. Among other aspects, this would facilitate comparison among surveys and provide an integrated view of the Agriculture and Rural sector.