Other

Testing and adjusting for attrition in Household Panel Data

author(s) Baulch, B.. - Quisumbing, A.
from UK Foreign, Commonwealth & Development Office (FCDO)
permalink open
themes
regions
  • Southern Asia
  • Asia
  • World
citation

Toolkit Note. Chronic Poverty Research Centre, London, UK, 12 pp

Publications and datasets provided via GARDIAN

Abstract

This note describes the use of a simple procedure to correct for attrition due to observables in household panel survey: inverse probability weights. The procedure involves estimating two probit regressions, one with and one without variables that are significantly associated with attrition, and using the ratio of predicted probabilities from these regressions to reweight the observations. The procedure is illustrated in Stata using data from part of the CPRC-DATA-IFPRI panel in rural Bangladesh

Open original