Variance estimates of poverty and inequality measures in Albania

The objective of the research reported in this paper is to contribute to the development of methodology and practical tools for computing sampling errors and design effects for complex statistics based on complex sampling designs, specifically sampling error of measures of poverty and inequality. It is taken as given that for the "typical" social surveys, based on reasonably large samples but with complex designs, the applicability of at least two broad approaches is generally well established in the literature-namely, the approaches based on Taylor linearization, and on resampling such as Jackknife Repeated Replication (JRR). This research has concentrated on elaborating the necessary details and developing software for their practical application by researchers who are not necessarily experts in the field of complex variance estimation.