In the current investigation, we used a single case alternating treatments design to evaluate the effects of a robot prompter on the social interaction of three pairs of children with autism spectrum disorder (ASD). Findings suggest that three participants produced more utterances and conversational turns when a robot prompter was present, and two when it was absent. One participant produced more utterances when the robot was not present, but more conversational turns when the robot was present. We discuss limitations to the current study and recommendations for future research.