Mind the mode: lessons from a web survey on household finances

Surveys on household income and wealth are generally carried out through personal interviews. In recent years, however, the spread of surveys using the Internet as a means of data collection has increased, both for economic reasons and for the speed with which the collected data are available. However, there are no studies on the effects of using this tool in surveys on household income and wealth. This paper contributes to fill this gap by illustrating the results of an experimental survey conducted in 2016 by the Bank of Italy in collaboration with the Italian national statistical institute (ISTAT). The quality of the information collected is assessed by comparison with tax statistics and with the aggregate evidence inferable from the interim survey conducted through personal interviews. The paper focuses on coverage, non-response and measurement errors. Overall, the results show that the web-based mode can be a valid alternative means of collecting qualitative data and of gathering information on less sensitive sources of income (such as employment and retirement income). In order to reduce the bias associated with the use of this instrument, it is essential, however, to have auxiliary information on the entire selected sample.