Recent debates within the autism advocacy community have raised difficult questions about who can credibly act as a representative of a particular population and what responsibilities that role entails. We attempt to answer these questions by defending a set of evaluative criteria that can be used to assess the legitimacy of advocacy organizations and other nonelectoral representatives. With these criteria in hand, we identify a form of misrepresentation common but not unique to autism advocacy, which we refer to as partial representation. Partial representation occurs when an actor claims to represent a particular group of people but appropriately engages with only a subset of that group. After highlighting symbolic and substantive harms associated with partial representation, we propose several strategies for overcoming it.