Modeling Multiple Adoption Decisions in a Joint Framework

Abstract

Abstract A multinomial probit (MNP) model is applied to the modeling of adoption decisions by farmers facing multiple technologies which can be adopted in various combinations. This model allows for full investigation of the interactions between decisions to adopt or not adopt several technologies. Estimation is carried out in a Bayesian framework employing Gibbs sampling to circumvent past difficulties encountered in maximum likelihood estimation of the MNP model. The model is estimated for a sample of U.S. apple growers with four possible sustainable production technology bundles. The results show that adoption decisions are most significantly influenced by off‐farm labor supply.