Intensive early intervention (EI) conducted in university clinics and model programs has been shown to dramatically improve symptoms and functioning in young children with autism spectrum disorders (ASD). Most children, however, receive EI in community programs. Little is known about community practices and the outcomes for children with ASD in these programs. The few studies of children receiving EI in community practice report that children make much smaller gains than observed in studies of university-based programs. This difference may result from differences in the quality of the intervention, the strategies used, or the children that participate in community as opposed to research programs. Research is needed that measures both the type and quality of EI in real-world settings and the associated outcomes, and that includes diverse samples representative of the range of backgrounds of children with ASD. Treatment studies for young children with ASD in the community (and in the clinic) also must address the question of which intervention works best for which child. Even in studies of model programs, the results can vary between children, and the overall beneficial effect observed is often due to large gains made by a small subset of participants. Little research has systematically examined the relative benefits of different interventions for children with different clinical profiles, which makes deciding between interventions and placements difficult for parents and providers. The aims of this project are to: 1) examine the type and quality of EI delivered in the community; 2) compare the effectiveness of different EI programs; and 3) investigate child characteristics associated with outcomes in each program. The investigators will compare outcomes in inclusive, mixed-disability, and autism-only community-based placements in a diverse, urban EI system. They will measure type and quality of intervention and child characteristics. The results will guide placement decisions and treatment improvements in real-world settings, thereby improving access to appropriate EI placements. The results also will lay the groundwork for a study of the match between child characteristics and community-based EI programs, which will promote the development of effective interventions that are more readily translated into community practice.