The purpose of the paper is to analyze the implications of using different methods to imput missing income data from the perspective of structure and class analysis. In particular, the interest is to describe and apply different methods and to examine the similarities and differences between them. In this way, it is expected to contribute to the problem of the lack of response in questions about income in household surveys. The methods considered are: Assignment of median income value according to social class and Multiple correspondence analysis and assignment of median income value according to clusters.