Statistical Matching Using Propensity Scores: Theory and Application to the Levy Institute Measure of Economic Well-Being

This paper summarizes the background, type, logic, and working procedure of the statistical matching used in the Levy Institute Measure of Economic Wellbeing (LIMEW) project to combine the various data sets used to produce the synthetic data set with which the LIMEW is constructed. We use the match between the 2001 Survey of Consumer Finances and Annual Demographic Survey of Current Population Survey data sets to demonstrate the procedure and results of the matching. Challenges facing the use of this technique, such as the distribution of weights, are discussed in the conclusion.