This paper is aimed at evaluating the incidence of measurement error on the main variables collected in the Bank of Italy’s Survey of Household Income and Wealth (SHIW). The results are especially relevant to researchers using the data for economic analysis, since they need to take data quality into account. Moreover, a thorough knowledge of the problems affecting the survey gives indications for improvements in its design and implementation. Where time-invariant variables are concerned, measurement error is studied by assessing the degree of inconsistency of answers given by panel households in subsequent survey waves. In the case of quantities that have an actual variation in time, such as income or wealth, the Heise (1969) model is applied; if data from at least three waves are available, we can separate the true dynamics from the noise of measurement error, under assumptions that are fairly mild. The essay also touches upon the role of fieldwork conditions, interviewer and respondent features in the determination of data quality.
Published in 2008 in: Review of Income and Wealth , v. 54, 3, pp. 466-493
