摘要：We analyze the non-pecuniary aspects of some crop biotechnologies taken from three farm-level surveys. We focus particularly on the phenomen on of part-whole bias, which is the empirical finding that the sum of the stated part-worths (the value of each nonpecuniary characteristic) is greater than the stated total value of all the non-pecuniary characteristics. We analyze the empirical evidence of part-whole bias in the surveys, while decomposing it to further understand the phenomenon and to rescale the stated values of the non-pecuniary characteristics in the surveys. We find for all three surveys that the degree to which part-worths should be rescaled is about 60 percent.
