Based on a randomized admission lottery in hundreds of local sites, the HSIS provides a platform for measuring Head Start's impacts--both short term and long term--on a nationally representative sample of participants without selection bias, providing internally and externally valid impact estimates for the overall Head Start participant population and for subsets of participants of particular policy interest. The proposed presentation will describe the study's path-breaking design and innovative analytic methods and share what the study team has learned about the sustained effects of Head Start as children complete 3rd grade. It will also explain the methods and sample used to determine these impacts. (author abstract)