Borges, M.R.S., Mendes, S., Motta, C.L.R.: Improving Meetings by identifying informal roles played by participants. Proc. of the 7th International Conference on Computer Supported Cooperative Work in Design - CSCWD´2002, Rio de Janeiro, Brazil, September 2002, pp. 368-373. Improving meetings by identifying informal roles played by participants Marcos R S Borges Sueli Mendes Claudia L. R. Motta Núcleo de Computação Eletrônica & Instituto de Matemática Universidade Federal do Rio de Janeiro Rio de Janeiro, Brazil mborges@nce.ufrj.br smendes@nce.ufrj.br claudiam@nce.ufrj.br Abstract The efficiency of a meeting has a lot to do with the attitudes the participants have towards the meeting goals. The outcome of a meeting is very dependent on how the meeting participants behave, i.e., how they assume expected roles. We can select participants based on their ability to play the desired roles prior to the meeting, or we can try to examine later the participants’ behavior by analyzing their interventions during the meeting. This paper discusses this later approach and suggests a semantics network analysis to map interventions onto roles, defined as essential to a good team performance. 1. Introduction Most people don’t like to spend time in meetings. To them, meetings always seem to last too long and accomplish too little. However, most participants and especially the meeting coordinator do very little to improve the effectiveness of the meetings. They don’t prepare themselves, don’t set an agenda, don’t follow-up the meeting’s decisions, and most importantly, they don’t run it well. As a result, many calls for a meeting bring the feeling of time wasting and worthless [10, 11]. The advent of meetingware has brought many benefits to the efficiency of meetings. Electronic meetings allow for asynchronism and parallelism, reducing the amount of time people spend in meetings and generating other benefits, such as automatic meeting memory [13]. However, no matter what technological support is provided, one still need the right people to produce the best possible outcomes [2]. Not only the right individuals but also, and most importantly, the best team. Participants with similar skills, ideas and attitudes will very seldom produce a good outcome [8]. Belbin [4] defines a team role as “a tendency to behave, contribute and interrelate with others in a particular way”. According to him, a work, a meeting included, carried out by a team would only be fully successful if the participants assume the essential roles. Belbin defines nine possible roles that a person can assume in a meeting. To a specific person, some roles are natural, some can be taken up, if necessary, and some are very hard to incarnate. In order to produce a good job, a team should consist of people able to play these roles, as naturally as possible. Belbin uses psychometric tests to relate observed team behavior to measured psychological traits and then construct balanced teams based on the identification of team’s roles [17]. Belbin’s approach is based on the use of test results to define the best team constitution before the meeting occurs. Differently from Belbin, our approach is to conclude about the team’s behavior by analyzing participants’ interventions after the meeting takes place. Although our approach does not prevent the occurrence of bad meetings, it signalizes to participants that a change of attitudes or addition of new members might turn the meeting more effective. In this work we use team roles as the framework for analyzing the behavior of meeting participants based on their interventions as recorded by an electronic meeting support system. We try to identify the characteristics of each intervention and relate them to team roles. Participants interact with each other through an electronic meeting system based on the IBIS argumentation model [5]. The intervention’s analysis uses the IBIS categories as a first hint to classify the type of statement produced by a participant. Then it uses a semantics network technique to identify the attitude embedded in the statement and map to one or more of Belbin’s roles. Our main goal is to identify missing roles and to guide further actions from the meeting’s coordinators. We claim that meetings can be improved when using this diagnosis to add other participants to the meeting, to review the agenda or to conduct the discussion to another direction. The rest of the paper is divided as follows. Section 2 describes in details the motivation of this work. It describes the problems of meetings, how electronic meetings have addressed them and those that remain unsolved. Section 3 describes Belbin’s roles and extracts from these roles the characteristics that best map interventions to them. Section 4 presents our approach to the process that maps interventions to roles. In Section 5 we illustrate the whole process with an example. Finally, Section 6 concludes the paper. 368