38 In addition, regarding Income Redistribution Survey, Gini coefficients before equivalence were different from the released values (Figure 3-3-1 carried in the above), but this is considered to have been attributed to the fact that the samples including replies judged to be irrational from the way of replying to items normally surveyed (such as Matsuura (2002) and Oshio (2005)) were excluded from the analysis. For the details of such method of additional treatment of individual data, refer to Appended Note 3-7.