The sampling design of the 2020 wave of the Italian Survey on Household Income and Wealth has been revised by exploiting newly available information on households' income and debt from register data. The redesign has led to a reduction in the bias of estimators of household income and wealth, but it has made it difficult to compare the results with those obtained in previous waves. This paper discusses three different weighting systems for taking the redesign into account.
The considered weighting strategies provide adjustments to sampling weights capable of making the distribution of income and debt in 2020 as similar as possible to that one would have obtained in the absence of a revision of the design. The methods discussed, although based on different assumptions, lead to similar and robust results.
