Although the predictive ability of ECT has been well demonstrated across a wide
range of contexts including marketing, electronic commerce, and information systems,
it has been the subject of several debates. Bhattacherjee (2001) summarizes these
debates well: (1) ECT ignores potential changes in consumers’ expectation after their
consumption experience, because post-purchase expectation may be different from
their pre-purchase expectation; (2) there are varying and conflicting conceptualizations
of the satisfaction construct in initial studies of ECT; and (3) conceptualization of
expectation also differs across ECT studies. Based on ECT and its limitations,
Bhattacherjee (2001) proposes a theoretical framework (namely a Post-Acceptance
Model) in the context of IS continuance. He argues that the proposed model mainly
focuses on post-acceptance variances because the effects of any pre-acceptance
variables are already captured within the confirmation and satisfaction constructs. The
proposed model includes ex post expectation, which is represented by perceived
usefulness as a post-consumption expectation (Bhattacherjee 2001). However, a
limitation of the model and a future direction for model extension, as Bhattacherjee
(2001) pointed out, is the dynamic interrelationships such as changes in the level of
expectations across pre- and post-phases that are not captured and tested. The first part
of Fig. 1 presents constructs and relationships in ECT. Note that the variables ranging
from expectation to repurchase intention form a sequential process and all constructs in
ECT except expectation are post-consumption variables. The second part of Fig. 1
shows associations among constructs in Post-Acceptance Model (PAM).
While PAM proposes several salient determinants of IS continuance intention
such as satisfaction and post-expectation (i.e., perceived usefulness) as beliefs
weighted with an evaluation of previous transactions, a couple of logical limitations
of the model are identified. First, satisfaction is an output of a consumer’s judgment
resulting from experiences and observations of previous transaction performance
(Oliver 1993; Oliver 1999). Perceived usefulness as a post-consumption expectation
is a salient belief consistently influencing future IS continuance intention. It is a
logical sequence to propose that a judgment of previous transaction (i.e.,
satisfaction) affects the level of anticipation of future performance (i.e., perceived
usefulness). Therefore, instead of the causality direction from perceived usefulness
to satisfaction proposed by PAM, it is more logically sound to propose the direction
of causality as the other way around (i.e., from satisfaction to perceived usefulness).
In addition, as Bhattacherjee (2001) discussed, the effects of any pre-acceptance
variables are already captured within the confirmation and satisfaction constructs.
Confirmation, in turn, influences the level of satisfaction positively and negatively
as a realization of the previous expectation. As an outcome of subjective judgment
of previous experience, satisfaction affects the level of perceived usefulness. Thus,
the causality relationship between confirmation and post-expectation (i.e., perceived
usefulness) is redundant in some sense.
3 Theory development and hypotheses
Synthesizing the importance of consumer trust and satisfaction from social
exchange theory along with two background theories (ECT and PAM), a theoretical
An investigation of the effect of online consumer trust 223
123
评论0