The Reliability of Matches in the 2002-2004 Vietnam Household Living Standards Survey Panel

Tracking households and individuals over time is important for a variety of research and policy questions. Exploring the validity of matched household and individuals within a dataset is a necessary step for ensuring the reliability of analysis designed to address such questions. This paper examines the quality of matching in the household panel between the 2002 and 2004 Vietnam Household Living Standards Surveys (VHLSS). Of the original 4,476 matches suggested in the household datasets, 429 matches are found to be incorrect, almost ten percent of total matches. Revised matches are suggested for 402 of these mismatches. Two simple applications illustrate the potential problems associated with analysis conducted using a poorly matched panel. The original panel overestimates the frequency of changes in household size. It also leads to biased estimates of per capita consumption growth, as it overestimates growth for poor households and underestimates growth for rich households.