Economic analysis of agricultural supply response frequently requires that county crop production data be combined A random coefficient regression model is used to test for aggregation bias under alternative US Department of Agnculture grouping schemes The criterion for aggregation is similarity across counties in yield growth rate, taken as a proxy for land quality Comparison of Crop Reportmg Districts (CRD's) and Major Land Resource Areas for Illinois supports the use of CRD's, the form in which supply data are often most readily available
