A Sample Selection Model of Consumer Adoption of Computer Banking

Abstract

We develop a bivariate probit selection model of consumer access and adoption that accounts for the cross equation correlations of the errors. The Survey of Consumers, collected by the University of Michigan, is the database used to estimate the model. We find a significant cross equation correlation between consumers’ perceived access and their use of computer banking. Based on our results, the bivariate selection model provides asymptotically more efficient estimates than does a single equation model because the bivariate selection model accounts for the sample selection bias associated with access. In addition, the bivariate selection model has a higher percentage of correctly predicted adopters than does the corresponding single equation (univariate) model.