Information Technology Reference
In-Depth Information
and normative beliefs, and including general fac-
tors such as personal innovativeness, seems to be
appropriate for describing technology adoption
decisions. The areas where results were not as
predicted, or where the paths changed from the
first to second time periods, bear particular atten-
tion. First, perceived behavioral control exerted a
positive influence on intentions at Time 2, but not
Time 1. In addition, we noted that the responses to
the perceived behavioral control items increased
significantly from Time 1 to Time 2. Keeping in
mind that one of the PBC items referred to having
the “… resources, the knowledge, and the ability
to make effective use of Access,” this suggests
that as the respondents gained experience with
the software, they gained more confidence in their
ability to control their decisions to use it.
This finding is not completely consistent with
Taylor and Todd (1995b), who noted a positive
influence of PBC on intentions for both inexperi-
enced and experienced users. The influence was
much stronger for experienced users, with path
coefficients of 0.16 for inexperienced and 0.50 for
experienced (Taylor & Todd, 1995b). In addition,
Taylor and Todd (1995b) tested their model within
a different context (use of a computer resource
center), which could explain the discrepancy in
results. In our study, respondents were required
to use Access for a course project. Even though
our questions focused on long-term, rather than
short-term intentions, it is possible that the man-
datory nature of the course project interfered
with their perceptions of control, and resulted in
intentions at Time 1 that were more based on the
course requirement than on their future plans. If
this were so, it is possible that there was both a
negative and positive influence at play, and these
cancelled each other.
Social factors did not exert an influence on
intentions at either time period. Keeping in mind
that the use of the software “in the future” would
be completely voluntary for the respondents, these
results are consistent with those of Venkatesh and
Davis (2000), who observed an influence of social
norms in mandatory, but not voluntary, settings.
We observed another interesting finding with
respect to the hypothesized influence of social
factors on perceived EOU. Recall that the scores
on the EOU scale did not change significantly
from Time 1 to Time 2, while the increase in the
scores on the social factors scale was statistically
significant. In addition, we hypothesized three
other factors as influencing EOU: self-efficacy,
personal innovativeness, and perceived behavioral
control. At Time 1, self-efficacy, personal innova-
tiveness, and perceived behavioral control all had
positive influences on EOU, while the path from
social factors to EOU was negative. At Time 2,
only the influences of self-efficacy and perceived
behavioral control were statistically significant.
There were at least two additional observa-
tions worth noting. First, computer self-efficacy
provided a strong, positive influence on PBC,
EOU, Affect, and PU (at Time 2). These results
show a much stronger role for self-efficacy than
what Taylor and Todd (1995a) and Venkatesh et
al. (2003) hypothesized in their models. In addi-
tion, CSE exerted a strong (indirect) influence on
intentions to use technology. Second, personal
innovativeness was also shown to exert an influ-
ence on intentions, both directly and indirectly
through CSE and EOU (at Time 1).
impliCations For researCh
and praCtiCe
One limitation of the study was the use of student
subjects, which limits the generalizability to some
extent (Compeau, Marcolin, & Kelley, 2001). This
provides an opportunity for future research, in that
it would be very useful to replicate the study in an
applied field setting with knowledge workers who
are being asked to adopt and use a new informa-
tion technology. A second limitation was the use
of a limited number of items (generally three) to
measure many of the constructs in the model.
Future research should include the development
Search WWH ::
Custom Search