Anonymised survey data are distributed for research purposes only. The use of the data is under full and exclusive responsibility of the author and may not involve the Bank of Italy in any way.
Authors who refer in their publications to results obtained with the survey data are required to cite the source (Bank of Italy, Survey on Household Income and Wealth) and to provide the version of the historical or annual archive used.
In order to keep the bibliography of the survey up-to-date, the author must send the complete reference of his/her research work based on the data to firstname.lastname@example.org.
The information required for data usage (the questionnaires of the latest surveys, the names of the variables, notes on data use, etc.) is given in the documentation.
All the data are available in three different formats: ASCII (‘comma separated’ CSV files), SAS and STATA (version 7 and above). Data are compressed in archive named: INDxx_format.zip (annual database) and STORICO_ format.zip (historical database), where xx denotes the last two digits of the survey year and format indicates whether the file is ASCII, SAS or STATA.
Statistics on the developments over time of the main economic indicators are available as Excel files in the statistical tables published in the section Survey results.
Version 11.3 of the historical database is updated and includes a set of new variables (the previous versions of the archive are available on request).
In order to improve sample representativeness of some population groups, in the 2020 survey the traditional sampling design in use since the 1989 edition was partly amended. The new design improves the quality of the estimators produced for economic analysis, but it makes comparison with previous waves more difficult. In order to overcome this limitation, two different weighting schemes were devised: one to be used for cross-section analysis, which includes the PESOFIT (sampling weight) and PESOFIT2 (population weight) variables in the CARCOM20 annual database and, for calculating sample variance, the PWTx variables in the PESIJACK20 database; one to be used for the purposes of historical comparison, which includes the PESO (sampling weight) and PESOPOP (population weight) variables in the PESO historical database and, for calculating sample variance, the PWTx variables in the PESIJACK database.
Furthermore, for prior editions (up to 2016) the sample weights included in the annual databases may differ from the corresponding weights included in the historical database. This is because the weights included in the annual databases are not revised, while the ones included in the historical database are aligned to the demographic statistics on the Italian population published by Istat (e.g. reconstructions between censuses) when the data are revised. These revisions may cause some discrepancies between the statistics computed using the current version of the Historical Database and those referring to the previous versions.