Mean-Covariance Structure Models in Economic Research: an Application to a Lending Program for Development in Burkina Faso

As applied development economic models become more sophisticated, they include increasingly complex conceptual variables. Due to data collection limitations, accurate proxies and continuous variables are often unavailable. A Mean and Covariance Structure model (MECOSA) is offered as a useful methodology for the incorporation of latent variables with metric, censored metric, dichotomous and ordinal indicators. As an example, conceptual variables (including borrower homogeneity and the domino effect) presented in the Besley and Coate (1995) group lending repayment game were specified as latent variables with non-metric indicators. Data from 140 groups from a group lending program in Burkina Faso were used to demonstrate the application and interpretation of MECOSA.