Exploring Quarterly Agricultural Survey Questionnaire Version Reduction Scenarios

摘要

The United States Department of Agricultures National Agricultural Statistics Service (NASS) conducts the Agricultural Survey Program (ASP), which consists of crop/stocks and livestock surveys. The Quarterly Agricultural Survey (QAS) questionnaire serves as a primary data collection instrument for the Agencys estimates of seasonal and annual crop production, supplies, and grain storage, which are used by the farming industry for both short and long term planning. The QAS is administered in all 50 states and collects data on 31 different crops and stocks, in varying combinations and frequencies throughout the year depending on the state. Such variation allegedly reduces respondent burden; however, it greatly increases the complexity of the survey administration process. Hierarchical clustering was done to investigate the potential of creating 20 regional QAS versions. Such an approach, if implemented operationally, would reduce the number of QAS versions by 60 percent (50 to 20). The research explored further clustering the QAS into only five regional versions, which, if implemented, would reduce the number of QAS versions by 90 percent (50 to 5).