DT 19 - 24 Imputación de ingresos del hogar en la Encuesta de Uso del Tiempo de Uruguay 2021-2022. Documento metodológico.

This methodological paper presents a strategy for imputing household income in Uruguay's 2021 Time Use Survey (TUS) using information from the Continuous Household Survey (CHS) from the same period. The need arises because the 2021 TUS, unlike previous editions, did not include questions about income but instead incorporated variables to construct the Socioeconomic Index (SEI).

Two imputation methods were evaluated: the first estimates independent regressions for each SEI decile in the CHS and then predicts income in the TUS; the second applies the 2 Multiple Imputation by Chained Equations (MICE) algorithm. Both methods use sociodemographic variables common to both surveys, including household head characteristics, household composition, and housing characteristics.

Results indicate that the SEI decile regression method produces better results, both in terms of average income and household income structure, especially in the middle and upper deciles of the distribution.

Keywords: Time use, income imputation, household surveys