COMPUTER SELF-EFFICACY - Human-Computer Interaction and Management Information Systems: Foundations - page 225

this context, and demonstrate how two specific differences (goal orientation and self-awareness)
influence the development of CSE, and thus other learning outcomes.
NEW TECHNOLOGY INTRODUCTION
Kelley et al. (1999, 2000) concentrate on the formation of specific computer self-efficacy in the con-
text of post-system conversion of new enterprise systems. From a theoretical standpoint, the degree
of behavioral change and learning required by complex computer technology such as enterprise sys-
tems makes them an ideal context for studying the formation of computer self-efficacy during the
learning process. The implementation of complex computer technology creates an environment that
is generally characterized by users as unpredictable, uncontrollable, and uncertain. Enterprise sys-
tem users are motivated to gain a realistic causal understanding of their changing computer envi-
ronment and to explain important computer-related performance events in order to predict their own
behavioral actions, such as usage patterns and control events, in their new computer environments.
Hence, the formation of CSE relies on a complex process of sense-making (i.e., attribution theory of
motivation), which includes interpretation and attribution activities by individuals. During this
sense-making process, individuals are attempting to identify the causes of actions and behaviors in
order to determine how they should respond to the new enterprise system during present and future
performance activities. Thus, users of recently implemented enterprise systems need to understand,
and make sense of, events and behaviors in their radically changing work environments before they
will effectively learn, adopt, and use this technology.
According to Bandura (1997), most motivation is cognitively generated through the exercise of
forethought or retrospective reasoning. People “form beliefs about what they can do, they antici-
pate likely positive and negative outcomes of different pursuits, and they set goals for themselves
and plan courses of action designed to realize valued future outcomes and avoid aversive ones” (p.
122). Outcome expectancies and cognized goals represent the key anticipatory mechanisms while
causal attributions of past performance represent the key retrospective mechanism.
Kelley et al. (1999, 2000) focused on one specific type of motivational processes—causal attribu-
tions from the theoretical perspective of Weiner's (1985) attribution theory of motivation. The basic
premise of Weiner's (1986) attribution model is that individuals' causal attributions of successful and
failed performance influence future expectancies and affect, which in turn influence future behavior.
These attributions may pertain to ability, effort, task difficulty, luck, mood, and help/hindrance from
other individuals and differ along the dimensions of locus of causality (i.e., internal/external to attrib-
uter), stability (i.e., constant/varied over time), and controllability (i.e., changeable/unchangeable).
The temporal sequence of events for attributional judgments begins with a performance out-
come that is interpreted by the individual as either successful or failed. Individuals then utilize
available information (e.g., performance history and performance of others) to infer explanations
specific to the events that caused the performance outcomes. The causal attributions of perfor-
mance determine future behaviors by providing the basis for subsequent actions to either continue
or discontinue the behavior.
The Kelley et al. (1999) study extends attribution theory by broadening the net of causal attribu-
tions that users attribute to computer performance and the related causal dimensions. They reported
that enterprise system users attributed their successful computer-related performance to willingness
to change applications, effort, and persistence. In contrast, enterprise system users ascribed their
failed computer-related performance to lack of computer training, lack of computer support, and
applications that were difficult to use. Of interest, post-survey interviews with enterprise system