The purpose of the proposed project is to investigate patterns, factors, mechanisms, and outcomes of Head Start/Early Head Start staff turnover using a range of advanced analytic techniques such as survival analysis, machine learning, Multilevel Cox Regression Model, and Multilevel Survival Structural Equation Modeling.  The study will involve secondary analyses of the Educare Learning Network National Evaluation longitudinal data to address questions about staff turnover.  As a subcontractor, FPG will be involved in assisting with any required data sharing agreements, creating custom datasets, and assisting with interpretation and dissemination of findings.