Supporting Distributed Groups with Group Support Systems - End User Computing Challenges and Technologies: Emerging Tools and Applications - page 229

Table 1. Subjects' backgrounds

subjects technology and training

There were 212 subjects in 47 groups in this
experiment. The subjects were recruited from
universities in the New York area. All subjects
were enrolled in undergraduate or graduate-level
MIS classes. During the recruiting session, the
subjects filled out preexperiment questionnaires;
were assigned to groups of five; and were sched-
uled for a one-hour, face-to-face training session.
After the training session, groups were assigned
randomly to experimental conditions. In a few
groups, however, the experiment was conducted
with three or four subjects due to dropouts during
the training session. The subjects' backgrounds
are summarized in Table 1.
Nunamaker, Briggs, Mittleman, Vogel, and
Balthazard (1996-1997) argue that established
groups of managers perform far better than the
student subjects used in most GSS experiment
studies. An advantage of using student groups,
however, is the accessibility to a larger pool of
subjects with little variances in their ages, educa-
tion, or business experience that would enhance
the statistical power in testing hypotheses. The
use of student subjects can be justified, because
when training is given to ad hoc groups before
an experiment, ad hoc groups seem to perform
as well as established groups (Mennecke, Hoffer,
& Wayne, 1992).

Electronic Information Exchange System 2 (EIES
2) was used for this experiment and modified to add
experiment-specific procedures and rules. EIES 2
is one of the major GSS research tools (Fjermestad
& Hiltz, 1998-1999) and is used frequently in
conducting asynchronous experiments for DGSS
research. It is similar to many group communica-
tion support systems now available for use through
the Internet. To minimize the problems associated
with student subjects, a week-long asynchronous
online training session with EIES 2 was given to
all subjects, beginning with a one-hour, face-to-
face session. All the subjects in a training group
were assigned to the same experimental group.
During training, all subjects were introduced to
the system's features and experimental formats.
Each group then selected a group leader, who
would serve as the group leader if the group were
assigned to the with-a-group-leader condition in
the experiment.

administration of the experimental
procedure

The experiment continued for two weeks. The
experimental procedures were constructed by
arranging discussion items, and the details are
summarized in Table 2. There was no face-to-face
session during the experiment. Each group mem-