Information Technology Reference
In-Depth Information
and testing of more appropriate scales for each
of the constructs.
The results of our analysis have several im-
plications for research on technology adoption,
as well as for organizational practice. The most
important implication for research is to reinforce
the arguments of authors such as Agarwal and
Karahanna (2000) and Plouffe et al. (2001) who
call for richer models of technology adoption.
Moreover, the results suggest that integration,
as well as richness, is of value to improving our
understanding of an individual's technology
adoption choice. Our integrative model shows
multiple mechanisms through which personal
innovativeness, self-efficacy, and social factors
inf luence technology adoption choices, and
adds to our understanding of how judgments of
perceived usefulness and perceived ease of use
are formed.
The results also show the influence of general
factors on specific software beliefs, suggesting
a degree of generality in perceptions relating to
computers. Bandura (1997) and others (e.g., Agar-
wal et al., 2000) argue persuasively for the need
to match self-efficacy judgments to the specific
task. This makes sense from the standpoint of
maximizing prediction, yet our results show that
these general influences can also influence beliefs
about specific software packages. Further study
of the generalizability of self-efficacy perceptions
(following the work of Agarwal et al., 2000) would
be valuable in building this understanding.
In addition, the strong influence of personal
innovativeness on self-efficacy and ease of use
perceptions (as well as on future intentions) sug-
gests that measuring personal innovativeness and
self-efficacy perceptions could help in developing
more effective training programs prior to the
introduction of new information technologies.
For example, knowing that a group of workers
scored highly on personal innovativeness would
suggest that less time and effort would be needed
to ensure they had positive beliefs about the ease
of use of a new information system. Further, the
relatively small influence of PBC at Time 1, fol-
lowed by the medium influence at Time 2, suggests
that the influence of PBC increases as users gain
experience. This suggests that managers should
ensure that potential users perceive they have
adequate access to resources (including training)
after they have had some initial experience with
the technology, and not just when it is introduced
to them.
Finally, the results confirm the importance
of incorporating experience into models of tech-
nology acceptance. Several authors have shown
changes in technology adoption models, and we
confirm their findings. We further show that it is
specific, more than general, measures that tend to
change with experience. The conceptualization of
experience is challenging however. As we noted
earlier, experience partly reflects exposure to the
tool and partly reflects the skills and abilities that
one gains through using a technology. Experience
also probably reflects habit to some extent. Our
research findings, including previous authors and
those in this study, do not clearly differentiate
between these types of effects. Nonetheless, it
seems reasonable that there might be different
sorts of influences depending on the extent to
which experience reflects habit, skill, or simply
exposure. Thus, we believe it is important for
future research to more fully examine the con-
ceptualization of experience and its influence in
technology adoption models.
ConClusion
In summary, our results provide support for an
extended model based on the decomposed theory
of planned behavior. They confirm existing find-
ings within the technology adoption stream, but
also show the possibility of a more holistic and
integrative approach to our models. Such an ap-
proach allows for the inclusion of less instrumental
beliefs (e.g., personal innovativeness with IT) as
influences on technology adoption, demonstrates
Search WWH ::
Custom Search