Hi dear R-helpers, Happy new year. I have estimated Gini index variance in the case of complex survey (stratify and with two stage sampling). I use the survey package. I use replication (bootstrap, and Jacknife) and linearisation according to Binder (1993) article David A. Binder and Milorad Kovaˇcevi´c(1993), « Estimating some measures of income inequality from survey data : An application of estimating equation approach », Proceedings of the ASA Survey Research Methods, 550-555. But I found different results. According to my experience, result of linearisation are more accurate. Is it an explanation to this ? Is bootstrap appropriate for all situations or there are drawback? Sincerly. Justin BEM BP 1917 Yaoundé Tél (237) 99597295 (237) 22040246 _____________________________________________________________________________ [[alternative HTML version deleted]]