Joseph E. McGrath

Universtiy of Illinois
Psychology Department
603 E. Daniel Street
Champaign, IL 61820

Phone: (217) 244-5876
Fax: (217) 333-4921
Email: jmcgrath@s.psych.uiuc.edu


SUMMARY OF MY NSF-IRI-ITO WORK
HOW DO COMPUTERS AFFECT COLLABORATIVE WORK IN GROUPS?

J. E. McGrath, University of Illinois

I am a social psychologist who for a long time has been studying small groups, especially in organizational contexts. For the past decade I have been working, along with a number of colleagues, on a sequence of three projects, under sponsorship of NSF - IRI - ITO, dealing with how electronic technology, especially computer conference systems, affect collaborative work in groups.



PROJECT ONE:
LITERATURE INTEGRATION - IRI 89-05640:
"Technological Enhancements and the Flow of Work in Groups"

In the first project, we systematically reviewed and tried to integrate the burgeoning research literature on "technologically enhanced" work in groups. That covered a wide range of areas -- email, computer conferences, closed circuit video conferences, and so on. But most of the work is on collaborative work via computers -- either synchronous collaboration as in computer online conferences or asynchronous collaboration as in email, bulletin boards, and many other forms. A lot of the results of the integrative review are available in:
McGrath & Hollingshead, 1995, Groups Interacting with Technology. Newbury Park CA: SAGE.



PROJECT TWO:
A LONGITUDINAL COMPARISON OF
COMPUTER AND FACE TO FACE GROUPS IR 91-07040:
"Time and Technology in Work Groups"

The second project centered around a large scale longitudinal (14 weeks) study -- technically, an "experimental simulation" -- comparing a set of small (3 and 4 person) work groups working face to face, versus a set of otherwise comparable groups working via a synchronous computer conference system. The groups worked for 2 hours each week. The participates were advanced undergraduates in a course on "the social psychology of organizations". They were formed into work teams as the workshop or laboratory portion of that course. They were asked to consider themselves a consultant team; each week they were asked to solve a problem given them by a "client organization". They worked on problems of different types each week, all projects more or less germane to organizational issues and to the material of the course. They also had one task common across weeks. Each week, all participants were asked to write an essay on how their work that week related to the material in the course. Then, they were asked to write an essay jointly as a group. Both individual and group essays were scored as part of the individual's grade in the course. [Thus, we presume, they were motivated to do well on the essays, whether or not they were motivated on the other tasks].

Members of the computer groups (CMC) worked in a lab with 16 terminals. Members of a given group were assigned to non-adjacent but networked terminals. The terminal screen, in addition to a continuous display of major control keystrokes and of work time remaining, had a box for typing and then sending messages, and a large message box for display of all messages sent by any member of that group. Members could scroll through the message box to reread past messages at will. Members of face to face groups (FTF) sat around a table and discussed their task, with some group member(s) writing down their solution on the task forms.

In that study, we carried out several important experimental manipulations. For example, we switched each group to the other medium (i.e., FTF to CMC, and CMC to FTF) for two weeks in the middle of the time period. Later on, we switched one member out of each group into another group for two weeks.

We refer to this study as JEMCO1. A number of studies resulting from JEMCO1 are published, together, in a special issue of the journal Small Group Research [vol. 24 No. 3 1993.


PROJECT THREE:
A REPLICATION WITH SOME IMPROVEMENTS IRI 93-10099:
"Technology, Experience, and Change in Work Groups"

The third project has centered around a quasi-replication of the previous study, which we called JEMCO2. It was a quasi- replication because we changed some important things. The main changes were that, half way through (in week 8), we switched all participants to the other medium, recomposed new groups, and left them in those new groups and that other medium for the remainder of the study. Therefore in that study we had more groups but for a shorter time (in Jemco1 we had 22 groups for 14 weeks each; in Jemco2 we had about 60 groups for seven weeks each). Moreover, each participant took part in a group using each medium. We also had a change in the software available for the computer groups (described later), which turned out to be very important. A number of studies resulting from JEMCO2 are published as a special double issue of CSCW: Computer Supported Cooperative Work, Vol. 4 March 1996.

We are now in the third and final year of that grant. Empirical results from our program include: information about alternative functions of computers in groups; evidence about how experience and changing conditions affects group task performance and patterns of member participation, both with and without technology; and a pattern of positive and negative effects of computer-mediated vs face-to-face technology. Our focus now is on integrating: (a) the extensive empirical evidence from Jemco1, Jemco2, and a number of smaller experiments in our program, with (b) a general theory of groups as complex, adaptive, dynamic systems, using (c) a computational model of group formation and development as a major tool in that integration.


Return to ITO Workshop Abstracts

Return to ITO Workshop Home Page