As the development and use of automated systems for collaborative work grows, the need for a better understanding of these systems becomes more important. Our focus is on one type of system, a Group Support System (GSS), and in particular, on one important aspect of a GSS, anonymity. A conceptual framework for the study of anonymity in a GSS is presented, which describes the general classes of variables and their relationships. These variables include the factors that influence anonymity in a GSS, types of anonymity and the effects of anonymity on a message sender, receiver, group process and outcome. Each of these variables is discussed with working propositions presented for important group process and outcome measures. The objectives of this paper are to highlight the importance and complexity of anonymity, to act as a guide for empirical investigations of anonymity, and to influence future GSS development and use.