Barry J. Fishman and Louis M. Gomez
The University of Michigan and Northwestern University
This paper explores the relationship between CMC tool use and classroom activity cycles. Results of case studies of students in two high schools using e- mail, Usenet news, and a multimedia groupware notebook are presented. Findings indicate that teacher designs for classroom activity relate strongly to the frequency with which different tools are used.
Keywords --computer networks, research in technologically-mediated communication, design and interface issues.
To date, relatively few studies have attempted to understand the role of collaborative communication technologies in classrooms. There are some notable exceptions: Schwab, Hart-Landsberg, Reder, and Abel ( 1992) studied how e-mail impacted the communication practices of middle school teachers. Levin, Kim and Riel ( 1989), and Levin, Waugh, Chung, and Miyake ( 1992) reported on calendar-based cycles in e-mail communications involving school students.
This paper reports on part of a larger program of research ( Fishman, 1996) that tracked student tool use behaviors and daily activities in thirteen classes taught by six teachers. These classrooms were equipped with (by current school standards) high-end Internet connectivity and computer-mediated communication (CMC) tools. These tools included e-mail, Usenet news, and an asynchronous multimedia tool called the CoVis Collaboratory Notebook. The Collaboratory Notebook, described at CSCL '95 ( O'Neill, Edelson, Gomez & D'Amico, 1995), is designed especially to support science inquiry.
Levin and his colleagues studied activity cycles in classrooms where electronic mail, a cornerstone CMC technology, was utilized. Research reports by Levin, et al. ( 1992; 1989) characterized the activity cycles of classrooms that participated in the Intercultural Learning Network ( Levin & Cohen, 1985) and AT&T Learning Circles ( Riel, 1989). They found that the ebb and flow of e-mail use in these environments corresponded to the daily, weekly, semesterly, and yearly cycles of the classroom. Over the course of a semester Levin, et al., found that communication activity would slowly build to a peak, and then fall away completely when everyone left for Christmas vacation. While this patterns seems obvious upon reflection, the point is made that this "calendar effect" needs to be of critical concern to designers from environments other than K-12 who may not take details of the academic calendar into account.
Levin, et al. ( 1992; 1989) only considered so called "calendar effects," and did not study additional effects of the task environment. A richer perspective on the activity cycle as it relates to tool use can be determined with a focus on the nature of the tasks that comprise project cycles, and how collaborative systems can in turn be instrumental to the successful completion of those tasks. Furthermore, Levin and his colleagues studied classrooms where e-mail was the only tool available, and only used for special purposes. The classrooms in this study have a rich "ecology" of CMC tools, and students and teachers may use them for a variety of tasks, both work-related and social in nature.
In classrooms that participated in this study, activity cycles correspond to the time it takes to initiate, pursue, and complete science projects. Project cycles can be very short--just a day or two-- or very long, lasting an entire semester. Common elements of all project cycles are a teacher introduction, a period when students seek information and data sources, a period when students begin to shape a final product, and a presentation and evaluation phase at the end. This cycle is similar to a model of research collaboration in white-collar environments that was described by Kraut, Galegher, and Egido ( 1987). In their model, Kraut, et al., divide projects into three major phases: initiation, execution, and public presentation. It is important to understand the different phases of the activity cycle in the classroom because in each phase there are unique demands placed upon CMC tools that are characteristic to the work being pursued. In the current study, one would expect to see patterns in CMC tool use that correspond to particular stages of the activity cycle. It was predicted that there would