Measurement errors in survey data and the estimation of poverty and inequality indices

Abstract

This paper firstly provides simple tools for evaluating the incidence of measurement error affecting the main variables collected in surveys on consumption. The assessment is carried out on two surveys that provide both diary and panel data. Diary data can be employed to obtain reliability coefficients for time-invariant variables. When variables vary over time, an estimation of the incidence of measurement error on the total variance can be obtained by applying models that allow the decomposition of observed variability into true dynamics and noise. Evaluations are also conducted on the basis of the internal consistency criterion. Finally, some methods for estimating the impacts of measurement errors on poverty and inequality analysis are also discussed.