The paper presents strategies for conducting research helpful to community instruction for students with autism. Research issues in setting variables (including heterogeneous/homogeneous groupings and instructional environments) are addressed as well as the use of the discrete trial format (including prompts, responses, and consequences). (CL)