This report provides advice to policymakers and school leaders on the use of assessment centres as part of a teacher selection approach. It discusses the relationship between assessment centre scores prior to joining teaching, and teacher effectiveness over a six-year period. It draws from various stages of a wider research project, the overall findings of which may be summarised as follows: (1) Assessment centre approaches may be useful in some forms of teacher selection; (2) Classroom simulation in the form of a teaching activity was the best predictor of future classroom practice; (3) Assessment centre scores are more likely to predict classroom effectiveness earlier in a teacher's career; and (4) Where lesson observation (and classroom simulation) is included in selection processes, using qualified teachers as observers can provide reliable scores with predictive ability.