Supporting Distributed Groups with Group Support Systems - End User Computing Challenges and Technologies: Emerging Tools and Applications - page 226

Figure 1. Research design

Independent Variables Dependent Variables

hypotheses development
decision Quality

Perceived Decision Quality

H2a: Groups with a leader will perceive that
their decisions are better than groups without
a leader.
H2b: Parallel communication groups will perceive
that their decisions are better than sequential
communication groups.

GSS research has shown clearly that the commu-
nication mode has a positive impact on a group's
decision quality. The impact of leadership in pre-
vious GSS studies, however, is inconsistent. The
impact either was not measured (Ho & Raman,
1991; Lim et al., 1994), tested insignificant (George
et al., 1990), or reported significant (Hiltz et al.,
1991; Tan, Wei, & Lee-Partridge, 1999). In GSS
research, because most tasks chosen for studies
are preference tasks, decision quality generally
is measured using either the perceptions of a
panel of experts or the responses to a question-
naire. Perceived decision quality, however, rarely
measures decision quality itself. Tan, et al. (1999)
assert that perceived decision quality measures
decision confidence, which is the degree to which
group members are sure that they have arrived at
an appropriate group decision. The nature of the
experimental task in this study, however, allows
objective decision quality as well as perceived
decision quality to be measured.

Consensus

Consensus refers to the degree of support among
group members in synthesizing divergent and
mutually conflicting ideas during interaction.
The level of consensus indicates what happens
during group interaction. Consensus also mea-
sures the degree of the acceptance of a decision
and the commitment to it (Dess & Orieger, 1987;
McGrath, 1984) and the level of effective comple-
tion of preferred tasks (Tan et al., 1999). Thus,
when implementing a decision is more important
than reaching a correct decision, consensus as the
measure of the acceptance of a decision should
take precedence over objective decision quality
(Dickson, Lee-Partridge, & Robinson, 1993).
Leadership in GSS settings generally shows no
significant impact on consensus. Ho and Raman
(1991) indicate that a leader in an unsupported
or unstructured group has more influence on
consensus, but the leader's effectiveness may be
canceled out when another process structuring
mechanism is present (Hiltz et al., 1991). These
findings suggest that leadership may be more
important when a group needs to establish a
structure for interaction. Hiltz et al. (1991) also

Objective Decision Quality

H1a: Groups with a leader will make better deci-
sions than groups without a leader.
H1b: Parallel communication groups will make
better decisions than sequential communication
groups.