A Bayesian model to predict the success of the implementation of health and education innovations in school-centered programs

Health and education practitioners, evaluators, and researchers have little guidance to help them translate implementation research into meaningful implementation strategies. This article describes the development and testing of a model to help schools assess their likelihood of successfully implementing health education innovations. The model was developed using an integrative group process technique that captures experts qualitative and quantitative judgments as a subjective Bayesian probability model. The experts developed a measurable definition of successful implementation, identified eight factors containing 40 questions relevant for predicting successful implementation, and specified the diagnostic value of each of the factors. Internal validation showed a correlation of 0.92 between the model scores and the experts direct ratings of 100 hypothetical school profiles. Preliminary external and content validation have been conducted. Application of the model to planning, management, and evaluation of school-based innovations is discussed.