The field research provides baseline data about how small groups work on specific tasks. Further, the field data helps suggest the kind of groupware that might assist that kind of work. Tasks that mimic the key aspects of behavior in the field have been developed for use in the lab. These are used to explore the value of various forms of groupware support for both face-to- face and distributed work. Different forms of audio and video support are also investigated for distributed work.
A series of investigations are under way or completed related to these project goals. Here are some recent results.
Audio vs. video for distributed groups
There is by now an extensive literature that shows that small groups
working synchronously in separate locations on some kind of intellective
task show no advantage of having video communication in addition to high
quality audio. The principle outcome measure in such studies is task
performance. The results show that being able to see each other does not
seem to add value to the task outcome when good audio is available, though
group members report that they like having the video. We have found
similar results ourselves in an earlier study reported in Olson, Olson &
Meader (1995).
A recent book edited by Wilbur et al (1997) summarizes several decades of research in this area, including two chapters by us in which we described our results and offered a detailed framework for the comparison and integration of a number of empirical studies.
In examining this literature in relation to a conceptual framework that relates group and task characteristics to technology support (Olson & Olson, in press a), we reasoned that there were situations not yet examined where video might play a critical role in task performance. Specifically, one such situation would be where the communication among group participants is difficult. Having had extensive experience with foreign students, we expected that groups doing a task in which they were speaking with each other in a second language would offer such a situation.
We carried out a study in which pairs of people did a task in whcih they needed to share information about a work object. Half of the groups consisted of members who were both native speakers of English, and half were groups were the members were non-native speakers of English and had different native languages. Half of the groups had high quality audio and half had audio plus high quality video. Consistent with our expectations, we found that the task performance of native speakers did not differ for the audio vs. video conditions (replicating the standard finding in the literature), while the non-native groups who had video performed significantly better than the non-native groups with only audio. This is one of the first clear findings of a substantial advantage in task performance for video-mediated distributed groups. This confirmed a hypothesis from our developing theory, namely, that when communication is stressed, video matters.
Comparison of groups in China and US
In collaboration with the Institute of Psychology of the Chinese Academy of
Science in Beijing we are replicating with Chinese subjects an earlier series
of NSF-sponsored studies carried out in the US (reported in Olson, Olson,
Storr¿sten & Carter, 1993). In step one of this project we have compared
US and Chinese subjects doing a standard design task in a face-to-face
setting. The 19 US groups were drawn from Olson et al (1993). These
were compared to 15 comparable groups of Chinese. We examined the
quality of the designs developed by these groups as well as the details of the
process by which they arrived at their designs, based on analysis of video
tapes. US and Chinese groups produced designs indistinguishable in
quality. Overall, the processes used in the two countries was quite similar.
Their discussions that focused on the design were similar in time allocation
and flow of activity. But they two kinds of groups did differ in how they
managed the synchronous session, with more formal meeting management
characteristic of the Chinese groups.
Following up with the research design in Olson et al (1993), we are currently cooperating with the Institute of Psychology to build a synchronous editor in Chinese that we will use to gather data about groups using a synchronous workspace tool in a face-to-face setting. The next step is then to look at distributed Chinese groups under conditions similar to Olson, Olson & Meader (1995).
The goal of this work is twofold: (a) to begin to look at the cultural specificity or universality of the line of work we have conducted previously, and (b) to assist colleagues at the Institute of Psychology in Beijing to learn about collaborative technologies and research methods for investigating them.
Comparison of synchronous drawing tools
We are conducting a study of the use of drawing tools for carrying out
synchronous graphical design. We have run three conditions, all with
groups of three in a face-to-face setting, using: (1) the LiveBoard, an
electronic whiteboard developed at Xerox PARC, (2) a workstation-based
drawing tool, and (3) an ordinary whiteboard. Not only will we compare
these three kinds of drawing support, we will also compare these more
visually based sessions with prior work we did with a tool that allowed easy
sharing of text only (Olson et al, 1993). We have completed the running of
subjects, and are in the midst of data analysis.
Investigation of trust in bargaining and negotiation
Most of our work so far has investigated cooperative tasks, in which the
members of a group are working together to accomplish a joint goal. We
are now beginning a new line of work in which we look at inherently
conflictful tasks, such as negotiation. In these tasks, the social dynamics
are quite different, and we would expect from our theoretical framework
(Olson & Olson, in press a) that the characteristics of tools that optimally
supported such tasks would differ from tools for cooperative tasks. We are
currently launching a set of field and laboratory studies on this topic, which
will constitute the major focus of our work on this project for the coming
year.
References