Developing an Appropriate Methodology for Farm Sampling in Order to Build a Representative Farm Database for Agricultural Policy Analysis Purpose. The Romanian Case.

摘要

The Common Agricultural Policy is foreseen to suffer important reforms after 2013. They may greatly affect the farm agricultural production because the way of accessing the direct payments will suffer important changes. In this context, impact studies at the farm level are very important. To do that, appropriate data about farms techniques, yields, costs, revenues and alike are needed. This paper investigates if the available Romanian statistical data are appropriate to be used in agricultural policy impact analyses. The lack of important information about Romanian farm’s techniques is evident in the official sources. Moreover, the aggregation biases can alter the agricultural policy impact analysis based upon simulative means. To overcome this, an appropriate methodology to collect data from farm level, at the North-West Romanian development region, was proposed. The farm population was firstly divided in homogenous strata. The first one deals with the farm type, the second one count for the geographical position and the third one with the farm size. Then, using official statistical data and the selective research methodology, a stratified random sample was proposed. It comprised 285 mixed cropping farms, 140 dairy and 200 sheep farms. The farmer’s professional organisation serves as a way of randomly identifying them into the survey. Individual data were collected based upon a face-to-face interview with the questionnaire as instrument.