ABSTRACT: As the use of computer-mediated systems for group work grows, the need for a better understanding of the impact of such systems becomes more important. Studies of non-computer-mediated group work have long been guided by mathematical models of group interaction and performance, but recent research suggests that these models do not fit the unique capabilities of computer-mediated systems. This article builds on Previous group research to create a mathematical model of computer-mediated group performance which is then tested using data from two prior studies. This testing enabled us to conclude that models of computer-mediated groups run counter to the many guiding models of non-computer-mediated groups. We therefore discuss the implications of these differences in relation to users of this technology and to future developmental and empirical research.
Key words and phrases: group interaction modeling, group process gains, group process losses, group decision support systems, idea generation