Abstract
Past test of differences in economic efficiency between small and large farms in developing countries used arbitrary criteria to divide samples of farms into the two size groups. Given the inferential danger from such arbitrary groupings, this paper presents a method for conducting the efficiency tests where the cut-off point between small and large farms is determined statistically. The test is applied to dryland food grain farming in Morocco. The estimated cut-off point was significantly different from the conventional one, but the efficiency test results were similar.