Bulletin, October/November 2006
Information Sharing of an Informatics Student Design Team
by Ammy Jiranida Phuwanartnurak
Ammy Phuwanartnurak is a doctoral candidate in the Information School at the University of Washington, Seattle. She can be reached at jiranida<at>u.washington.edu.
In the spring 2005 I studied the information sharing of an informatics student design team from a database class. The design team worked on designing and developing a small, but quite complex, information retrieval system as part of the class requirements. This article highlights important findings from the study in relation to finding patterns of information sharing in student teams. Cognitive work analysis (CWA) was used as a framework in collecting and analyzing data. Interviews and email logs were key methods for collecting data.
Data were analyzed with respect to the CWA framework. For this study, I chose to focus on the dimensions of CWA – work environment, work domain, organizational and social interaction, and the actor’s resources and values. The means-ends analysis template is a tool used in CWA to analyze complexity in the study context. Briefly, it is an abstraction hierarchy, ranging from the most abstract level (goal and constraints) to the most concrete level (physical resources). In this study, the means-ends analysis template was used as a tool to analyze work environment and work domain, while a social-organizational analysis was used to analyze the team’s organizational and social interaction and the actor’s resources and values. In addition to the means-ends analysis, email communications were analyzed to further understand the information sharing and social interaction of the team. The findings reveal the nature of information sharing as well as the factors that shape the information sharing practices of the student design team.
The Student Design Team
The student design team participating in the study was one of 13 student teams taking a database class. Each student team, selected by the instructor and a teaching assistant based on a skills profile, consisted of three to four students. The team members took this class primarily to meet the informatics program requirements and to learn about database design. The class was constrained by many factors such as school and the university requirements, instructor and students’ knowledge and information technologies. However, a major constraint pointed out by all team members was a schedule. Besides passing the class, students considered learning about the database and getting hands-on experience working with databases as priorities of this class.
The team consisted of three junior informatics students – Alan, Bill and Charles (pseudonyms). The goal of the team was to design and develop a detailed specification and a working prototype of an information system focusing on the evolution of cities and towns. The designed system would enable people to submit photographs and other media (such as audio files) of places so that people could see how a building, view, landscape feature or artifact has changed over time. The team had approximately 10 weeks to complete this design project, which was subject to a number of project requirements developed by the instructor. The main priorities of this project included finishing deliverables on time and developing a usable database system.
The Organization of the Team’s Work
Team members’ competence and the desire to share work load equally were the main criteria for role allocation. The task division was also somewhat constrained by project requirements specified by the instructor as outlined in the project description document. That is, the design and coding must be done individually by a designated lead person. At the planning stage, the work was divided fairly equally. For example, each member individually wrote about one third of the first draft of the project report. Later on, when the team started coding tasks, team members selected the part of the work with which they felt comfortable.
Although the team did not have a clear social-organizational structure, its structure could be best described as autocratic. Bill appeared to have more influence on how the work should be done than the other two members. Alan and Charles perceived Bill to be more knowledgeable, while Bill believed that he could accomplish this project by himself. Bill was often the first to start working on project deliverables, and he would tell others what they would (or should) do next. Alan and Charles often felt like Bill was a supervisor and they were employees.
All team members said that they would prefer a more democratic structure where all members participated equally in the work. They would like to communicate more and be more open with each other. However, the team did not make any attempt to communicate more openly. One reason could be that the team members all had other responsibilities such as other classes, individual homework and exams. Therefore, they did not have much time and effort to dedicate to this design project.
Team members described team culture as both laid-back and deadline-driven. No work was done until a few days before due dates. Although all team members said that they expected their team culture to change, they had done nothing to change it. Reviewing the team members’ comments, one could conclude that perhaps there was no change because the group project was worth only 30% of the total class grade and the available time to spend on the project and the team was reduced by the amount of work from other classes.
Alan and Charles were closer to each other than to Bill. Interaction and collaboration often happened between Alan and Charles, with the result that Alan and Charles usually worked closely together, while Bill seemed to prefer working independently. In addition, their working styles were different; Bill liked to work in the morning, while the other two members did not. Most of team communication was via email, although, interestingly, all team members said that they valued face-to-face meetings and would like to have regular group meetings. However, those meetings did not happen. Besides seeing each other in class, the team met only a few times over the course of the project.
Information Sharing Practices
The team shared information for collaboration and to complete a task. Their primary information sharing activities were in the forms of the draft report, codes, updated progress reports and technical knowledge. For example, Alan would send email messages to Bill and Charles to update them on his work. The team used email, Wiki and face-to-face meetings to share information.
Email. There were a total of 57 email messages sent among team members over the course of the project with Bill being the most active with 31 email messages and Charles the least active with five messages to the team. Bill initiated a large number of email communications.
Email communication did not happen until the fourth week, during which the first deliverable was due. There was no email between team members during weeks 5 and 7. The email communications started to pick up again in week 8 when the second deliverable was due. Clearly, information sharing via email often took place when the deliverables were due.
Wiki. Bill created a project wiki – a wiki is a type of website that allows users to create and edit Web page content very quickly and easily using any Web browser – so that the team could collaborate and share project-related information. Bill posted most of his work on the wiki while Alan and Charles never posted anything on the wiki.
Face-to-face meeting. All team members told me that they would like to have group meetings throughout the course of the project. However, the team could not schedule a weekly or bi-weekly meeting. Instead, they met only a few times at night before the deliverables were due. Meetings usually occurred a day before the deliverable due date, and they were more like work meetings in which team members were working individually on their parts. Most face-to-face collaboration happened during class time.
The study shows that the student design team tried to share information using several methods. The necessity for these various methods was due primarily to the time constraints faced by the students that made face-to-face information sharing difficult. In addition information sharing practices were affected by class and project requirements, individual characteristics and knowledge of available technologies. While they tried to use several technical methods for overcoming the lack of face time, these methods didn’t always work. For instance, Alan and Charles didn’t use the wiki because they were unfamiliar with the technology. As a result, most information sharing took place by email, which enabled the bulk of information sharing in the team.
Information sharing is essential to collaboration for any design project. Before conducting this study, I expected that information sharing would happen regularly during a student design project. In contrast, the findings show that the student design team rarely shared information. Given this situation, how can we support information sharing in student design teams? To promote information sharing in student design teams, instructors need to think about various dimensions – individual characteristics, social, time and technological influences – when developing a course curriculum and project requirements because, as shown in this study, all of these dimensions affect student information-sharing behaviors.
This article also points out that to understand information sharing in student design teams it is important to examine not only the information sharing itself but also the constraints that shape the sharing practices. This observation would seem to apply for studying professional design teams as well. Note that the constraints both restrict and enable the team actions. For example, as previously stated, the lack of jointly available time required the use of communications technologies such as email. Absent email, it would have been difficult for the team to communicate at all.
Cognitive work analysis is a useful framework for studying information sharing in design teams because it allows researchers to learn the constraints on how people work. Examining work and task alone provides a limited amount of information. Using the CWA framework to determine the constraints gives a more complete picture of why the work is done a certain way, not just what work is done. In the case of student design teams the information found in this study regarding the time constraints facing team members could be useful information for instructors, who can gain a better insight into how their students actually work on team projects. This new information might then inform the design of student assignments as well as instructor expectations regarding student behavior when working on assignments in teams.
Articles in this Issue
Information Sharing of an Informatics Student Design Team