Self-Efficacy: A Theoretical Analysis of Its Determinants and Malleability

Author(s): Marilyn E. Gist and Terence R. Mitchell

Source: The Academy of Management Review, Vol. 17, No. 2 (Apr., 1992), pp. 183-211
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/258770 .
Accessed: 09/05/2014 14:28

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy
of Management Review.

http://www.jstor.org

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
0 Academy of Management Review
1992, Vol. 17, No. 2, 183-211.

SELF-EFFICACY:A THEORETICAL
ANALYSISOF
ITS DETERMINANTSAND MALLEABILITY
MARILYNE. GIST
TERENCER. MITCHELL
University of Washington
The construct of self-efficacy has received increasing empirical atten-
tion in the organizational behavior literature. People who think they
can perform well on a task do better than those who think they will
fail. Differences in self-efficacy are associated with bona fide differ-
ences in skill level; however, efficacy perceptions also may be influ-
enced by differences in personality, motivation, and the task itself.
This article reviews theoretically the antecedent processes and infor-
mation cues involved in the formation of self-efficacy. A model of the
determinants of self-efficacy is proposed that enhances understand-
ing of both the complexity and malleability of the construct. Determi-
nants that facilitate the most immediate change in self-efficacy are
identified, and appropriate change strategies are highlighted. Impli-
cations and propositions pertaining to future research are discussed
at the end of the article.

A recent addition to the organizational research agenda is self-efficacy,

a person's estimate of his or her capacity to orchestrate performance on a
specific task. Empirical studies of self-efficacy have yielded several consis-
tent findings. For example, self-efficacy is associated with work-related per-
formance: life insurance sales (Barling & Beattie, 1983), faculty research
productivity (Taylor, Locke, Lee, & Gist, 1984), coping with difficult career-
related tasks (Stumpf, Brief, & Hartman, 1987), career choice (Lent, Brown,
& Larkin, 1987), learning and achievement (Campbell & Hackett, 1986;
Wood & Locke, 1987), and adaptability to new technology (Hill, Smith, &
Mann, 1987). Also, studies have indicated that some training methods can
enhance self-efficacy in the areas of self-management (Frayne & Latham,
1987), cognitive modeling (Gist, 1989), and behavioral modeling (Gist,
Schwoerer, & Rosen, 1989). Further, when self-efficacy is enhanced, atten-
dant increases in performance are noted (Gist, 1989; Gist et al., 1989).
Although these findings demonstrate the importance of self-efficacy for
predicting and improving work performance, much remains unclear about
the construct itself. The purpose of this article is to explore two major theo-
retical issues that hold implications for future research on self-efficacy.

This work is dedicated to the memory of Dr. Morris Rosenberg, recipient of distinguished
scholarship awards from the American Sociological Association and the University of Mary-
land. His teachings on the sociology of the self-concept and his 1979 book, Conceiving the Self,
have had significant influence on many of the ideas presented here. Appreciation is expressed
to Albert Bandura for his comments on an earlier version of this article.

183

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
184 Academy of Management Review April

First, there is a need to know more about how individuals form judg-
ments of efficacy. Though Bandura (1986) described four broad categories of
experience that contribute to efficacy judgments, greater specification is
needed to understand the information that is drawn from personal and work
experiences and utilized in the formation of self-efficacy. For example, mod-
eling is known to influence self-efficacy, and the symbolic coding process is
recognized as the mechanism by which individuals interpret the behavior
they observe in others (Decker, 1980). However, less is known about the
types of information, generated through observing others, that ultimately
determine self-efficacy. This article offers a classification schema for infor-
mation cues that influence perceived efficacy.
The second theoretical issue concerns the malleability of self-efficacy.
In part, the question of how self-efficacy can be changed (e.g., through
training) may be a question about how beliefs about abilities or motivation
may be changed. However, greater conceptualization is needed about the
plasticity of the determinants of self-efficacy: the specific causal factors that
are susceptible to change, the extent of probable change, and the practical
issues involved in facilitating change.
Comprehension of these issues requires some grounding in knowledge
of self-efficacy that is derived from allied disciplines of study. Even though
space does not permit an exhaustive review of this literature, an abbrevi-
ated summary is provided of the construct of self-efficacy and its theoretical
context. Following this summary are presentations of the theoretical issues
central to this article: the determinants of self-efficacy and the process of
change. Some propositions and implications for research are discussed at
the end of the article.

ON SELF-EFFICACY
PERSPECTIVE
Construct Definition Issues
Self-efficacy is a construct derived from social cognitive theory-a the-
ory positing a triadic reciprocal causation model in which behavior, cogni-
tions, and the environment all influence each other in a dynamic fashion
(Bandura, 1977, 1986). Wood and Bandura (1989a: 408) stated that "self-
efficacy refers to beliefs in one's capabilities to mobilize the motivation,
cognitive resources, and courses of action needed to meet given situational
demands."
Other work has highlighted three aspects of this definition (cf. Bandura,
1988a; Bandura & Wood, 1989; Wood & Bandura, 1989b). First, self-efficacy
is a comprehensive summary or judgment of perceived capability for per-
forming a specific task. In an organizational context, information derived
from the individual, the work task, and others in the work environment may
contribute to the comprehensive assessment of capability. Second, self-
efficacy is a dynamic construct. The efficacy judgment changes over time as
new information and experience are acquired (sometimes during actual
task performance). Third, efficacy beliefs involve a mobilization component;

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 185

self-efficacy reflects "a more complex and generative process involving the
construction and orchestration of adaptive performance to fit changing cir-
cumstances" (Bandura, personal communication, 1989). Thus, people who
have the same skills may perform differently based on their utilization,
combination, and sequencing of these skills in an evolving context.
Related constructs. Conceptual differentiation between self-efficacy
and similar constructs is important for further theoretical development on
self-efficacy. One of two constructs most frequently confused with self-
efficacy is self-esteem, although there are important differences between
the two. Self-esteem usually is considered to be a trait reflecting an indi-
vidual's characteristic, affective evaluation of the self (e.g., feelings of self-
worth or self-liking). By contrast, self-efficacy is a judgment about task ca-
pability that is not inherently evaluative. For example, a rocket scientist may
have very low self-efficacy pertaining to dancing, yet may decide on re-
flection that this is satisfactory and that it does not diminish his or her overall
evaluation and feelings about the self.
Brockner (1988) provided a thorough discussion of the distinctions be-
tween self-efficacy, self-esteem, and other self-related constructs. He also
discussed conceptualizations of global (i.e., evaluation of the total self) and
specific (i.e., situational or task-specific) self-esteem. By contrast, self-
efficacy always refers to task-specific capability. "Some researchers (for
example, Shrauger, 1972) have studied task-specific self-esteem by assess-
ing individuals' confidence about succeeding at a given task; this opera-
tional definition of task-specific self-esteem appears to be synonymous with
self-efficacy. Self-esteem typically refers to a global construct that taps in-
dividuals' self-evaluations (and not merely their confidence judgments)
across a wide variety of situations" (Brockner, 1988: 14).
Self-efficacy also has been likened to expectancy. Vroom's (1964) theory
that expectations influence action refers to two types of probabilistic esti-
mates: behavior-outcome relationships, such as effort to performance (Ex-
pectancy 1 or E1), and outcome-outcome contingencies, such as perfor-
mance to reward (Expectancy 2 or E2). E1 is considered more analogous to
self-efficacy; in fact, both constructs flow from similar theoretical notions that
expected performance outcomes depend heavily on the type of behaviors
an individual chooses to execute.
However, several conceptual and measurement distinctions between
self-efficacy and expectancy theory concepts have been articulated previ-
ously (cf. Bandura, 1984; Bandura, 1986; Gist, 1987;Locke, Frederick, Lee, &
Bobko, 1984). In general, conceptual distinctions between self-efficacy and
E1 suggest that self-efficacy may represent a more comprehensive formu-
lation of the rationale underlying the expectancy theory construct. For ex-
ample, the most frequent conceptualization and empirical assessment of EF
is in terms of the relationship between effort and performance (Mitchell,
1974). By contrast, Bandura asserted not only that self-efficacy subsumes
variables typically not included in E1 (such as mood), but also that selfef
percepts of efficacy are not simply inert predictors of future behavior" (1984:

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
186 Academy of Management Review April

242). Though both constructs involve forethought, self-efficacy is viewed as

having generative capability: it influences thought patterns, emotional re-
actions, and the orchestration of performance through the adroit use of
subskills, ingenuity, resourcefulness, and so forth (Bandura, 1984, 1986).
Measures in use also contribute to the divergence (or convergence) of
these constructs. Locke and his colleagues (1984) articulated a number of
differences typically found between effort-performance and self-efficacy
measures (e.g., self-efficacy measures often distinguish between magni-
tude and strength, whereas effort-performance measures typically do not).
As is the case with differences between self-esteem and self-efficacy, oper-
ational definitions of E1 sometimes approach in similarity those typically
used for self-efficacy (i.e., assessing expectancies across a range of perfor-
mance levels rather than for one assigned performance goal, and assessing
for immediately subsequent performance-cf. Ilgen, Nebeker, & Pritchard,
1981). However, self-efficacy generally encompasses a broader range of
predictors of a performance level for a specific task, whereas expectancy
focuses on a narrower behavioral predictor of overall performance on a job.
Relevance to self-regulation. Self-efficacy is one of several cognitive
processes frequently considered in self-regulation (a comprehensive pro-
cess of cognitive, individual determination of behavior). Convergent evi-
dence for the importance of self-regulation in motivation and performance
is provided by several streams of organizational research. These include,
but are not limited to, goal setting (cf. Locke & Latham, 1990), control theory
(see Klein, 1989, for a review of various control theory perspectives), and an
integrative model of motivation and cognitive abilities (Kanfer & Ackerman,
1989).
Even though a complete review of the role of self-efficacy in self-
regulation is beyond the scope of this article, several consistent findings
deserve mention. First, self-efficacy has been shown to influence both goal
level and goal commitment (Locke et al., 1984; Taylor et al., 1984). Second,
self-efficacy influences an individual's initial choice of activities and tasks
and his or her coping efforts while engaged in these tasks (Lent et al., 1987;
Stumpf et al., 1987). "People process, weigh, and integrate diverse sources
of information concerning their capabilities, and they regulate their behav-
ioral choices and effort expenditure accordingly" (Klein, 1989: 167). Finally,
self-efficacy influences the interpretation of feedback (Silver, Mitchell, &
Gist, 1991) and affective reactions to the task (Gist et al., 1989; Kanfer &
Ackerman, 1989), which have an impact on the self-regulatory processes
that influence subsequent performance (Bandura, 1986; Kanfer & Acker-
man, 1989).
In sum, self-efficacy is an important motivational construct. It influences
individual choices, goals, emotional reactions, effort, coping, and persis-
tence. Self-efficacy also changes as a result of learning, experience, and
feedback. To clarify further the construct of self-efficacy (and to consider its
determinants), brief attention should be given to how the construct is as-
sessed and the factors influencing that assessment.

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 187

Measurement Issues

The traditional measurement of self-efficacy (discussed by Bandura,

1984, 1986) requires that an individual respond dichotomously (yes or no) to
whether he or she is capable of performing at several specific levels on a
specific task (the sum of positive responses is the magnitude of self-efficacy).
For each affirmative response, confidence is then rated on a scale that
ranges from 1 or 10 (quite uncertain) to 100 (quite certain) at 1- or 10-point
intervals, respectively (cf. Gist, 1989;Gist et al., 1989). The sum of confidence
ratings is the strength of self-efficacy. A variation is that strength may be
measured by eliciting and summing the confidence ratings for all perfor-
mance levels (as opposed to only those for which there is an affirmative
response-cf. Locke et al., 1984). Whichever method is used, scores are
then correlated with performance across subjects. Magnitude and strength
can account for independent sources of variation in performance, so both
may be used. However, strength seems to be preferred for analyses be-
cause of its wider variance. On some occasions, Likert-type scales have
been used which simply ask how well the person thinks he or she can do on
the task: that scale score is then correlated with performance (Bandura,
1977; Hill et al., 1987; Schunk, 1983, 1984).
An important question arising from the measurement protocol is
"whether subjects can accurately predict their own behavior" (Eastman &
Marzillier, 1984: 224). The methodological issue inherent in this question is
the construct validity of self-efficacy. Based on the literature cited previ-
ously, the predictive validity of self-efficacy is well established, and predic-
tive validity is part of overall construct validity (Cook & Campbell, 1979).
However, self-efficacy is unique from a construct validity perspective.
In many behavioral research situations, construct validity is established by
showing that different measures of the same construct are highly correlated
(Campbell & Fisk, 1959). When measuring self-efficacy, an individual is
asked to predict performance, yet the criterion to which self-efficacy should
be related most is also performance. Thus, predictive validity for self-
efficacy is conceptually similar to the way in which construct validity would
be assessed, and support for predictive validity can be construed as partial
support for construct validity (Landy, 1986). In this sense, support for the
theory is very similar to the type of support generated for the Fishbein and
Ajzen (1975) and Ajzen and Fishbein (1980) model of attitudes, in which a
very specific intention to engage in a very specific behavior is correlated
with the demonstrated behavior. When the correlations are high, then the
support is strong and unambiguous; it supports both the measure and the
theory.
Problems arise when the correlations, although supportive, are not
very strong (e.g., Locke et al., 1984; Stumpf et al., 1987; Taylor et al., 1984;
Wood & Locke, 1987). A moderate or low correlation can mean that (a)
self-efficacy is not strongly related to performance on this particular task
(e.g., a task in which luck plays a large part, such as gambling), (b) the

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
188 Academy of Management Review April

measure of self-efficacy was not very good (i.e., weak differentiation of

performance levels), (c) there was difficulty in assessing self-efficacy accu-
rately (e.g., a task that was new to the subject), or (d) the theory is wrong.
A number of factors can degrade this self-efficacy-performance rela-
tionship. For example, complex tasks (cf. Campbell, 1988; Wood, 1986) re-
quire the subject to estimate numerous skill and motivational parameters
(Bandura 1988b). The complexity of this cognitive activity may increase the
error of assessment. Second, the time between self-efficacy assessment and
criterion assessment should decrease the predictability of self-efficacy, as
demonstrated by Wood and Locke (1987). Third, inaccurate or ambiguous
feedback can degrade the self-efficacy -* performance relationship. Fourth,
the less well defined the performance levels are on the task, the lower the
correlation (Eastman & Marzillier, 1984). And fifth, the less experience peo-
ple have with a task, the less accurate their prediction of performance
(Bandura, 1988b; Kanfer, 1991). The problem for the researcher is to deter-
mine whether low correlations between self-efficacy and performance are
due to the complexity of the task involved, when and how self-efficacy was
assessed, or the person's familiarity with the task.
Another implication for construct validity arises from examining the
measurement protocol. When people are asked to evaluate their "capabil-
ity" of performing a task at specific levels, what do they consider exactly?
This question is addressed in the next section. Given the foregoing meth-
odological discussion, the subsequent ideas in this article are based on the
assumption that the task for which self-efficacy is being assessed has rela-
tively well defined (i.e., unambiguous) performance levels. Also, although
self-efficacy is a dynamic construct, at any given point of measurement, a
person must consider a variety of information cues to form a judgment. The
following analysis of self-efficacy determinants depicts such a point in time.

OF SELF-EFFICACY
DETERMINANTS

Theoretical Analysis
Bandura suggested that four categories of experience are used in the
development of self-efficacy: enactive mastery (personal attainments), vi-
carious experience (modeling), verbal persuasion, and physiological
arousal (e.g., anxiety). Although these experiences influence efficacy per-
ceptions, it is the individual's cognitive appraisal and integration of these
experiences that ultimately determine self-efficacy (Bandura, 1982). Thus,
self-efficacy may be thought of as a superordinate judgment of performance
capability that is induced by the assimilation and integration of multiple
performance determinants.
Theoretical implications of these experiences and their attendant effi-
cacy perceptions have been articulated for organizational behavior and
human resource management (Gist, 1987). However, little attempt has been
made to identify and organize the specific information cues provided by the

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
 1992 Gist and Mitchell 189

four types of experience. Further, limited understanding exists as to how

individuals evaluate those cues in forming self-efficacy.
Figure 1 presents a simplified view of the process by which self-efficacy
is formed. Note that the major emphasis is placed on those judgments and
information categories that precede the efficacy assessment; the conse-
quences of efficacy beliefs (e.g., goals) are not the focal concern in this
figure. Three types of assessment processes appear to be involved in form-
ing self-efficacy. First, there is an analysis of task requirements. This anal-
ysis produces inferences about what it takes to perform at various levels. For
example, someone considering the task of making a market forecast may
consider the extent to which mathematical abilities (e.g., statistics) will be
required in order to perform well and how much time will be needed. This
task analysis should be most explicit when the task is fairly novel or when
it has been observed only. When the task has been performed personally
and frequently in the past, the individual is likely to rely more heavily on his
or her interpretation of the causes of previous performance levels.
Thus, a second form of analysis is typically involved in efficacy judg-
ments: an attributional analysis of experience. This analysis involves the
individual's judgments, or attributions, about why a particular performance
level occurred (e.g., "I've done good forecasts in the past which I attribute
to my math skills and/or my hard work"). People seek answers to the ques-
tion of why things happen, both out of a desire for mastery as well as out of
simple curiosity (cf. Kelley & Michela, 1980; Weiner, 1985). Kelley (1971: 22)

FIGURE1
A Model of Self-Efficacy-Performance Relationship
Analysis of Task
Requirements

Enactive Mastery sAttributional Etimation of Consequences of

Vicarious Experience _ A nalysis of Orchestration Self-Efficacy Performance
Verbal Persuasion ExAlrience Capacity (e.g., goal level, lm a
Physiological Arousal pe(Self-Efficacy) persistence)

t t+ I
I \ Assessment of / l
| a Personal and /
Situational Feedback
Resources/ I
l Constraints I l

I
l

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
190 Academy of Management Review April

asserted, "The attributor is not simply an attributor, a seeker after knowl-

edge: his latent goal is that of effective management of himself and his
environment." Although personal experiences may provide the strongest
information for attributional analysis, causal information also can be in-
ferred from experiences such as modeling or persuasion. For example,
through the symbolic coding process (cf. Decker, 1980), an individual may
deduce the relevant skills and behaviors used by a model in performing a
task, approximate the extent to which those skills are similar to his or her
own, and infer the amount of his or her own effort versus skill that would be
required to produce a comparable performance.
The assessment of task requirements and attributional analysis of ex-
perience provide some sense of what it will take to do well on the task in
terms of ability and motivational components, and in terms of the relative
contributions of these to performance. However, these two antecedent pro-
cesses appear to yield necessary but insufficient data in the formation of
self-efficacy. There remains an examination of self and setting by which the
individual assesses the availability of specific resources and constraints for
performing the task at various levels. This assessment requires consider-
ation of personal factors (e.g., skill level, anxiety, desire, available effort) as
well as situational factors (e.g., competing demands, distractions) that im-
pinge on future performance.
Though these three assessment processes are relatively independent,
progression through them may occur in an iterative manner, and the rela-
tive emphasis on each process may be influenced by the nature of the task
itself or by the extent of prior experience with the task. These assessment
processes yield interpretive data that are used in a summary-level judg-
ment process which defines self-efficacy: the estimation of orchestration
capacity. Indeed, Bandura (1988b) argued that self-appraisal is a process in
which different sources of information are weighted and integrated to form
self-efficacy, and that the relative weighting of information may vary across
domains of functioning and situational circumstances.
Information-Processing Issues
The theoretical analysis of the process by which self-efficacy is formed,
presented in Figure 1, can imply extensive cognitive activity or very limited
activity. Even though it is beyond the scope of this article to present a
detailed description of how various cues may be processed and integrated
(indeed, there are extensive possibilities), some issues about this processing
deserve attention.
Depth of processing. An important distinction is noted between a fairly
in-depth analysis (often called controlled processing) and a quicker, more
superficial judgment process (often called automatic). A number of authors
have focused on this distinction in areas such as attitude development and
change (Petty & Cacioppo, 1986), attitude strength (Fazio, 1986), task per-
ceptions (Zalesny & Ford, 1990), decision making (Mitchell & Beach, 1990),
mental health (Langer, 1989), organizational behaviors (Weiss & Ilgen,

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 191

1985), performance appraisal (Feldman, 1981), and group norms (Gersick &
Hackman, 1990).
There exist several theoretical descriptions of the systematic, detailed,
controlled, in-depth, and analytic approaches (Beach & Mitchell, 1978;
Einhorn & Hogarth, 1981; Feldman, 1981; Graik & Lockhart, 1972; Petty &
Cacioppo, 1986), as well as theories of judgment processes described as
automatic, intuitive, mindless, unanalytic, or schema driven (Chaiken,
1986; Feldman, 1981; Schneider & Schriffen, 1977; Taylor & Crocker, 1981).
Although these theories differ in some important ways in focus, there is
some agreement about the factors that prompt controlled versus automatic
processes to occur. Specifically, novelty, unexpected failure, salience or
importance of the task, and changes in major aspects of the self or task tend
to prompt more rigorous analysis. Also, with respect to more in-depth or
controlled judgments, methodological procedures have been developed
that demonstrate how multiple cues are weighted and evaluated. These
methods include the lens model (Beach, 1967; Brunswick, 1956; Slovic &
Lichtenstein, 1971) and functional measurement procedures (Anderson &
Zalinski, 1988).
It is suggested here that judgments about efficacy become more rou-
tinized and automatic as experience with a task increases. Drawing on the
previous literature, when tasks are novel, when changes have taken place
in the person or task which affect performance, or when the task is salient
or important to the individual, a more detailed analysis is likely. In these
situations, a person may assess, in depth, the task demands, the environ-
mental constraints and support, and his or her own attributes and feelings
when forming self-efficacy.
However, under routine circumstances, the individual is apt to refer
simply to his or her previous performance level and to utilize that level as
the primary determinant of self-efficacy. Between these two extremes may
be situations in which the person considers past performance, makes simple
attributions about the causes of past performance, and adjusts self-efficacy
accordingly, or invokes scripts or other cognitive schema (perhaps devel-
oped from vicarious experience or experience with other tasks) when judg-
ing efficacy.
Accuracy. The accuracy of self-efficacy judgments can be viewed in
various ways. Most global is the accuracy of the efficacy judgment itself.
This accuracy basically is reflected by the predictive validity of self-efficacy.
The higher the correlation, the greater its accuracy, but a number of sub-
ordinate accuracy judgments are important for overall accuracy. These are
estimates of (a) the amount of a particular resource that is needed to perform
well, (b) how much of an attribute or resource the individual has or the
situation provides, (c) how various resources and attributes contribute to
performance relative to other resources, and (d) the specific attributions that
are made about causes of performance.
Two major factors appear to influence the accuracy of these assess-
ments. First, as suggested previously, there is the individual's experience

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
192 Academy of Management Review April

with the task. As a person gains experience with a task, it is less likely that
detailed task and resource analyses will be done and it is more likely that
the person simply will use past performance and attributions about the
causes of that performance as major predictors of how he or she can per-
form. In general, past experience regarding the task should increase the
accuracy of self-efficacy to the extent that causal attributions about past
performance are accurate.
The second factor concerns personal and task stability. If personal char-
acteristics have changed in major ways or are currently undergoing
change, an individual may be less accurate in judging efficacy. Similarly,
if the task itself involves characteristics that change (e.g., working with a
new person may be different than working with a known co-worker), the
less stable the task attributes, the less accurate self-efficacy may be.
In summary, forming an efficacy judgment initially may require an
extensive analysis of the task and the person's own situational resources
and constraints. As experience with the task increases, shorthand evalua-
tions and stable attitudes about the task may evolve; these may be reflected
in the formation of self-efficacy through simpler processes of recalling past
performance level(s) and making attributions (perhaps aided by schema or
scripts, such as "I did a mediocre job last time, but then I never do well at
those types of tasks, so I'm sure I'll do poorly next time too"). Thus, the
judgment process, with repeated exposure to a task, moves from controlled
to automatic, and from weakly held, unstable and potentially less accurate
judgments to more strongly held, stable, and accurate ones. Indirect sup-
port for this is suggested by Bandura's (1982) identification of a hierarchy of
influence in forming self-efficacy, whereby mastery experiences (personal
attainments) are viewed as providing the strongest information for self-
efficacy, followed in decreasing order of influence by vicarious experience,
verbal persuasion, and physiological arousal. Accumulated personal ex-
perience with a task provides direct information about performance attain-
ments, whereas data from modeling, persuasion, and arousal are less di-
rect. Direct knowledge about capabilities should lead to more automatic
information processing and the formation of efficacy judgments that may be
more stable and accurate, compared to judgments formed from the indirect
information provided by modeling, persuasion, and arousal.

AN OVERVIEW
OF EFFICACYCUES
The following specification of self-efficacy determinants reflects an in-
tellectual debt to attribution theory in two distinct ways. First, one of the
antecedent processes involved in forming efficacy judgments has been de-
scribed as a causal attribution process. The resulting attributions become
cues for subsequent self-efficacy. Thus, some of the determinants of self-
efficacy are well-recognized attributional causes (i.e., effort, ability, luck,
task difficulty). These attributions are clearly distinct from efficacy beliefs;
whereas attributions are assessments about causes of past behavior, self-

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 193

efficacy pertains to future performance capability (cf. Silver et al., 1991).

However, attributions made through a causal analysis of the past are useful
in assessments of future capability. "The factors singled out by attribution
theory serve as conveyors of efficacy-related information that influence per-
formance attainments mainly by altering people's beliefs in their efficacy"
(Bandura, 1988a: 38). Specifically, self-efficacy mediates the attribution -*
performance relationship (Bandura, 1988a; Fosterling, 1985; Schunk &
Gunn, 1986).
The second contribution of attribution theory for understanding self-
efficacy determinants is derived from the work of Weiner (1985) and others
who provided distinctions (i.e., internal/external, stable/unstable, controlla-
ble/uncontrollable) that could be used to categorize attributions. The same
distinctions can be used to group the factors influencing efficacy judgments.
More specifically, a variety of factors that influence efficacy are external task
factors (e.g., group interdependence, distractions such as noise), whereas
other factors are internal (e.g., health, mood). Also, some of these factors
can either change or be changed readily (e.g., distractions and mood), but
other factors may be more resistant to change (e.g., group interdependence
or health). Finally, some factors are more or less under the control of the
person (e.g., effort), some are largely controlled by the organization (e.g.,
task resources), and some may be uncontrollable (e.g., weather, temporary
illness). Thus, the structural framework (but not the content) of attribution
theory can be helpful for organizing the types of cues that influence efficacy
assessments.
One final comment is needed regarding this overview. As mentioned
previously, efficacy judgments may require an analysis of a large number
of cues. External factors appear to affect self-efficacy indirectly through their
influence on internal variables, such as motivation, ability, performance
strategies, and so on. Some of these same internal factors (i.e., motivation,
ability, performance strategies) can be estimated directly through reflection
and internal focus. This point is important for later discussion of the ways in
which self-efficacy can be changed by methods that are designed to change
these external factors or internal assessments.
External Cues
The primary external cues pertain to the task itself, and they indirectly
influence self-efficacy through their evaluation by the individual. First, the
task attributes are important. Individual estimates of efficacy may include
considerations of the degree of interdependence (cf. Kiggundu, 1981) and
the amount of resources (i.e., material resources, time, and staff) required to
complete the task successfully (Bandura, 1988a). In an organizational con-
text, a lead programmer asked if he or she is capable of completing a large
software project within a certain schedule would consider factors such as
whether others who are working on interdependent subcomponents of the
program will complete their portions in time, whether user time is required
on the hardware to fully test the software, and so on. Evidence that task

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
194 Academy of Management Review April

attributes are determinants of self-efficacy is offered by Cervone (1985), who

demonstrated that when people were asked to focus on formidable aspects
of the task, their self-efficacy was lowered. Self-efficacy was increased by
focusing on doable aspects of the task.
Another external cue pertaining to any task is its complexity (Campbell,
1988; Wood, 1986). In the formation of self-efficacy, the important compo-
nents of complexity include the number of component parts involved in
completing the task, uncertainty (dynamic elements or probabilistic link-
ages), and the sequential or coordinative steps required to perform the task
successfully. Cervone and Peake (1986), for example, changed efficacy
judgments simply by manipulating (i.e., increasing or decreasing) the level
of possible performance attainments on a task.
The task's environment also may influence performance estimates. For
example, when a person performs a task that requires the application of
analytical skills in a task environment in which there are many distractions
(e.g., noise, interruptions), this environment may lower performance esti-
mates when compared with a task environment that is less distracting. Also,
the amount of risk or danger in the setting (whether physical or psycholog-
ical) may influence self-efficacy. Distractions and risk may increase anxiety,
which can reduce efficacy through thoughts of failure, physiological man-
ifestations of stress, and the reduction of coping mechanisms (Lazarus &
Folkman, 1984). Other environmental factors that may relate to self-efficacy
are physical conditions (e.g., weather) or the geographic setting (e.g., per-
formance of a negotiator at the home office versus performance at the op-
ponent's site).
Watching others (modeling) also may provide information about abili-
ties. "People partly judge their capabilities in comparison with others"
(Bandura, 1988c: 143). Modeling appears to be a particularly effective
means of providing information on "correct" performance strategies, be-
cause this information may not be available otherwise (even from past
performance). In addition, psychological strategies (such as coping with
boredom or anxiety, or persisting despite difficulty) can be evaluated by
observing others. Also, task familiarity and the observation of others pro-
vide knowledge about the relative contribution of ability and motivational
components (e.g., effort, persistence) required by the task (Kanfer & Acker-
man, 1989).
Information cues derived from external verbal persuasion may include
feedback or instruction about abilities. Bandura and Cervone (1986) dem-
onstrated that feedback information (in the form of a discrepancy between
performance and a personal standard or goal) can influence self-efficacy.
Of critical importance are the credibility, expertise, trustworthiness, and
prestige of the person doing the persuading (Bandura, 1977).
Other persuasive discussions (i.e., pure persuasion) may lack specific
information about the person's ability and focus primarily on convincing an
individual, through emotional or cognitive appeals (e.g., exhortations), that
he or she can perform a task at a certain level. Though such appeals may

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 195

result in more weakly held efficacy beliefs because less is learned about the
requirements of task performance, they clearly can have an impact on
efficacy judgments (Schunk, 1983, 1984). Arguments resting on pure per-
suasion may provide information about the amount of effort that an indi-
vidual should expend (perhaps altering goals, priorities, interest, or affect).
Internal Cues
Familiarity with a task through personal or vicarious experience will
provide information that is directly accessible. Of particular importance is
the adequacy of an individual's ability (knowledge and skills) and the ef-
fectiveness of various performance strategies that utilize these skills
(Bandura, 1988c; Earley, Connelly, & Ekegren, 1989). Different types of task
performance strategies may facilitate performance. These may include be-
havioral strategies (such as feedback-seeking or interpersonal negotiations),
analytical strategies (such as breaking the task into subparts for ease of
cognitive processing or identifying a simpler way of solving a problem than
is typically considered), and psychological strategies (such as persisting
despite difficulty, coping with boredom, or managing anxiety). Self-efficacy
will be determined, in part, by the individual's assessment of whether his or
her abilities and strategies are adequate, inferior, or superior for perfor-
mance at various task levels. Also relevant is whether the person believes
these abilities and strategies are either fixed and immutable (i.e., inborn
talent) for the task or can be acquired or improved through additional train-
ing and experience (Wood & Bandura, 1989a).
Individuals also make judgments about anticipated performance based
on how positively aroused (i.e., excited, enthusiastic) or negatively aroused
(i.e., fearful, anxious) they feel when confronted with a particular task
(Bandura, 1988a, 1988b). Arousal may be influenced partly by external vari-
ables discussed previously (e.g., distractions, risk). However, three other
factors may induce arousal. First, there is general physical condition. For
example, an existing ulcer may inhibit performance on tasks that a person
deems stressful. Second, there are certain personality factors that influence
arousal, subsequent performance, and probably self-efficacy. For example,
Type A personalities, recognized for having generally higher levels of psy-
chological arousal than Type Bs (Friedman & Rosenman, 1974), have been
shown to have higher self-efficacy and performance (Taylor et al., 1984). In
contrast, individuals with low self-esteem, who suffer from anxiety in many
task-performance situations, typically have an elevated self-attentional fo-
cus (i.e., concern about their anxieties and inadequacies) that detracts from
task performance (Brockner, 1979). Thus, it is expected that self-efficacy may
be depressed across a greater variety of tasks for individuals with low self-
esteem when compared to those who are high in self-esteem. Finally, a
person's immediate affect (or mood) may affect arousal and efficacy. Mood
can be determined by numerous events that are both task related (Russell &
McAuley, 1986) and not task related, such as incidents in a person's private
life, the kind of traffic the person encountered on the way to work, or the

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
196 Academy of Management Review April

time of day. Kavanagh and Bower (1985) demonstrated that a mood ma-
nipulation resulted in higher self-efficacy scores for a positive state (mood)
and lower self-efficacy scores for a negative state.
A Three-Dimensional Model of Efficacy Determinants
As discussed previously, the experiences of mastery, modeling, per-
suasion, and arousal (Bandura, 1986) are more complex than their labels
imply; each of these experiences contributes a variety of external and in-
ternal information cues that can influence self-efficacy. However, this con-
sideration examines only one dimension of attribution theory: the locus of
causality. Two other major dimensions which should be considered are the
variability (over time and occasions) and the controllability of the causal
influence. In Figure 2, we have presented a two X two table which orga-
nizes the cues according to their locus of causality and variability. Also, we
have indicated the extent to which these cues are associated with control-
lability.
In the low variability/external determinant category are factors pertain-
ing to the attributes of the task (e.g., interdependence and resources) and
the complexity of the task (e.g., difficulty, uncertainty). Low variability/
internal determinants refer to assessments of ability as well as stable dis-
positional attributes (e.g., physical condition, personality). The high vari-
ability/external cues are produced by the task environment (e.g., distrac-
tions, risk). Finally, the high variability/internal determinants include
current performance strategies, and motivation to exert effort (which is in-
fluenced by goals, priorities, interest, and mood).
The third distinction suggested by attribution theory is controllability
(Weiner, 1979). When making efficacy assessments, individuals also con-
sider whether they exercise control over the determinants (Bandura &
Wood, 1989). However, controllability can vary in several ways. First, some
factors are primarily under personal control (depicted as internal in Figure
2), and some are primarily under the control of others in the organization
(depicted as external in Figure 2). In many cases, individuals have, or
perceive themselves to have, little or no control or influence over external
factors. Thus, perceived control is likely to be higher over internal than
external factors.
A second way in which control can vary is based on the time horizon
between self-efficacy assessment and performance. Some factors can
change immediately (such as effort), whereas others may be changeable,
but only after a long period of time (such as ability or general physical
condition). Thus, perceived control may be higher over determinants that
are immediately variable than over those that are relatively more stable.
Finally, independent of the time available, some factors may be rela-
tively uncontrollable by either the organization or the individual (e.g., for-
tuitous resources arising from unanticipated profits, temporary physical ill-
ness). Thus, controllability is partly, though not entirely, independent of the
locus and stability of efficacy determinants shown in Figure 2. With the

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
 1992 Gist and Mitchell 197

FIGURE2
Determinants of Self-Efficacy
Locus of Determinant0

External Internal

Task Attributes Ability

Interdependence Knowledge
Resources Skills
L
Task Complexity General Physical Condition
0 Average fitness level
Number of General health
W component parts Skill variety

Sequential Personality
performance Type A/B
requirements Self-esteem
Variability
of
Determinantb
Interpersonal
Environment
Persuasion Performance Strategies
Feedback Behavioral
H Models Analytical
Psychological
Task Environment
Distractions Effort
G Risk/Danger Goals
Physical/Geographical Priorities
H setting Interest
Xr Affect (mood)

a
External determinants are primarily under the organization's control. Internal determinants
are primarily under personal control.

b
In general, greater control may be perceived over highly variable determinants than over
determinants low in variability. With the exception of spontaneous, luck-oriented factors
(such as temporary illness), the high variability internal determinants should correspond with
the highest perceived control.

exception of spontaneous, luck-oriented factors (temporary illness, etc.),

when individuals make attributions to internal, more variable causes, con-
trollability should be perceived as highest (Fosterling, 1985). When control
seems low, especially over causes that are likely to lead to poor perfor-
mance (e.g., poor health or risk), negative outcomes in the form of learned
helplessness or anxiety may occur (Mikulincer & Nizan, 1988; Seligman,

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
198 Academy of Management Review April

1975). Holding constant other variables, the greater the personal control, the
higher self-efficacy should be (Fosterling, 1985).
In sum, experiences such as mastery and modeling provide information
cues regarding efficacy that can be categorized using the distinctions of
externality, stability, and controllability provided by Weiner (1985). The
meaning of these cues (in terms of relative importance or weighting) ap-
pears to be inferred primarily from an attributional analysis of experience
when tasks are familiar. When tasks or personal/situational factors are new
or have changed significantly, the meaning of cues may be inferred from an
in-depth analysis of task requirements, coupled with an assessment of per-
sonal and situational resources/constraints. Weighted cues can then be pro-
cessed (cf. earlier references to theories and methodological procedures
involved in complex judgment formation) to estimate orchestration capacity
for performance (i.e., self-efficacy). In turn, self-efficacy affects performance
through behavioral choices such as goal level, effort, and persistence.
When researchers conceptualize performance information cues in terms of
their underlying dimensions (e.g., externality), this process facilitates an
understanding of how self-efficacy may be changed-an important topic
for improving motivation and performance.

IMPLICATIONS
FOR ENHANCINGSELF-EFFICACY

The significance of self-efficacy in predicting performance on work-

related tasks has been demonstrated clearly by the studies cited previously.
An important implication for researchers and practitioners is that an in-
crease in positive beliefs or a reduction of debilitating beliefs may lead to
higher task performance. Some organizational studies have reported posi-
tive posttest differences in mean self-efficacy and performance for partici-
pants who were exposed to various forms of self-management and model-
ing training (Frayne & Latham, 1987; Gist, 1989; Gist et al., 1989). Other
studies in the clinical and educational arenas have used techniques based
on operant principles (Andrews & Debus, 1978), modeling (Zoeller, Ma-
honey, & Weiner, 1983), information (Wilson & Linville, 1985), and persua-
sion (Schunk, 1984) to change beliefs, attributions, expectancies, and self-
efficacy (Bandura, 1977). However, the extent to which self-efficacy and
performance can be raised and the overall malleability of self-efficacy still
are unresolved issues. Little attempt has been made to understand system-
atically which determinants might be altered to yield the greatest change in
self-efficacy and which change strategies should be used. The model shown
in Figure 2 serves as a theoretical framework for conceptualizing the
change process.

Limits to Change

Based on the previous discussion, the degree of change in self-efficacy

should be influenced by four factors. These are the initial level of self-

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 199

efficacy (high or low), and the variability, locus, and controllability of the
determinants of self-efficacy.
Level effects. Within person analyses of self-efficacy increases should
be subject to ceiling effects when pretest self-efficacy is high. Regardless of
its accuracy (predictive validity), there is little room for positive change in
self-efficacy level. In contrast, low self-efficacy will be less subject to a ceil-
ing effect and can be increased contingent on the extent to which initial
efficacy perceptions were inaccurate.
When low self-efficacy is an inaccurately low assessment of perfor-
mance "capability," this perceptual error may result from incorrect assess-
ments of task or individual influences on performance (i.e., how much and
of what is required, how much the person has, and the relative contribu-
tions of each determinant). Positive information about these determinants
(e.g., reassurance about task resources or abilities, emphasis on the need
for concentration and effort) can lead to greater increases in self-efficacy
when it is low than when it is already high. Alternatively, low self-efficacy
may be an accurate reflection of performance capability. In this case, the
level of self-efficacy can be raised through verbal persuasion, but this in-
crease in self-efficacy probably will reduce its predictive validity and will be
of short duration if performance fails to improve.
Thus, two processes seem to account for increases in self-efficacy level
which lead to corresponding increases in performance. The first is direct
intervention to correct inaccurate and low perceptions of performance ca-
pability. Such misperceptions may be widespread, but they are by no
means the only cause of low self-efficacy. When low self-efficacy results
from an accurate assessment of low capability, self-efficacy is best in-
creased indirectly through an alteration in task and personal factors that are
directly related to performance levels (e.g., skill level, distractions).
Stability and locus of determinant. Much of the research on changing
self-efficacy has been conducted in clinical or educational settings (cf.
Bandura, 1986; Wilson & Linville, 1985). However, these findings may not
generalize completely to organizational contexts where the task demands
may be quite different. Using the model in Figure 2, differences in self-
efficacy between phobics (e.g., snake phobias) and nonphobics might arise
from the locus and variability of the determinants of self-efficacy. Regarding
phobias, the focal determinants may be primarily internal and highly vari-
able (e.g., effort exerted toward approaching the snake, a performance
strategy such as psychological coping with anxiety). The motor and intel-
lectual skills required for touching a snake are present in most individuals.
By contrast, many work tasks require knowledge and skills that must be
learned over time through extensive training. To the extent that ability is a
relatively important and less variable determinant of performance than
effort, the potential for generating immediate and accurate changes in self-
efficacy will be limited.
A further understanding of how self-efficacy can be changed is en-
hanced by attention to external determinants: the task and its environment.

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
200 Academy of Management Review April

Kanfer and Ackerman (1989) and Norman and Bobrow (1975) pointed out
that tasks vary in the extent to which performance is sensitive to the amount
and type of resources (e.g., effort, ability) expended. Some tasks are re-
source dependent, whereas other tasks are resource insensitive (Wickens,
1984). On the one hand, if the task is resource insensitive (i.e., ability
and effort are not strongly related to performance), then either the individ-
ual or the organization must change the task or its attributes (e.g., uncer-
tainty, risk) if self-efficacy is to be increased and subsequent performance
enhanced. On the other hand, if performance on the task is sensitive to
ability, effort, and persistence, then efforts to change self-efficacy (by
changing beliefs, information, and knowledge) may lead to performance
increases.
The degree of change that might be expected in self-efficacy is partly a
function of the variability and locus of its determinants; thus, individual
differences in the composition of self-efficacy must be considered when
interventions are designed to enhance it. As an illustration, using Figure 3,
consider the situation of two budget analysts facing intensive preparation of
divisional forecasts for the coming fiscal year. Assume that each individual
has an identical, accurate (i.e., very predictive), and low self-efficacy for the
task. Individual A perceives that the forecasting task is high in complexity
and that she has limited skills (technical knowledge) to apply to it. She
generally copes very well with the high pressures of her other work; thus,
she plans to work hard on the forecasting task, although she is troubled
about her lack of requisite task knowledge. In contrast, Individual B ac-
knowledges the high task complexity, but feels he has excellent skills for the
forecasting job. However, he is disinclined to work hard on it because of his
high volume of other work and his anticipation that one of the managers
from whom he needs input will be typically late in providing it.

FIGURE 3
Individual Differences in Composition of Self-Efficacy
Individual A Individual B

20%
Variable

70%
Variable

80% Stable

30% Stable

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 201

Because ability is a necessary, but not sufficient, requirement for per-

formance (i.e., a person can have high ability but intend to put forth no
effort), the relative contribution of ability and effort to perceived capability
(i.e., self-efficacy) and performance can vary widely across individuals for
the same task. Note that two people can have either different levels of a
particular factor (e.g., ability) or weight that factor differently in the extent to
which they perceive it contributing to their performance. Thus, the same
self-efficacy score can result from multiple combinations of weights and
values assigned during the estimation of orchestration capacity. In the ex-
ample above, the low self-efficacy of Individual B results primarily from the
heavy weights and low values applied to psychological coping and effort
(relatively variable determinants). Individual A's low self-efficacy results
primarily from the heavy weight and low value applied to her technical
knowledge (a more stable determinant). An intervention designed to en-
hance self-efficacy through training and providing task knowledge may
yield greater improvement in self-efficacy and performance for Individual A
than for Individual B. Similarly, interventions designed to increase inten-
tions to expend effort (e.g., goal setting, incentives) may increase self-
efficacy substantially for Individual B, but only marginally for Individual A.
Thus, a change in the mean level of self-efficacy may lead to a change
in the mean level of performance across individuals. However, it is the
highly variable, internal determinants that are most likely to lead to imme-
diate changes in self-efficacy and performance within individuals. Some
support for this is found in the educational literature where the greatest
amount of positive change in self-efficacy, effort, and persistence occurs
when people are led to believe their failures are due to lack of effort (Fos-
terling, 1985). Similarly, Anderson (1983) found that subjects who believed
task performance was determined by variable causes (e.g., effort, task strat-
egies) were less discouraged by failure than people who believed that the
causes of performance were stable (e.g., character dispositions and ability).
Controllability. The more that people believe that the causes of per-
formance are uncontrollable, the lower and more resistant to change will be
their self-efficacy. However, most of the determinants of efficacy are at least
partially under the control of the individual and the organization, although
the relative control will vary. As was mentioned, internal cues like effort
may be controlled directly by the individual, whereas external cues such as
task interdependence tend to be controlled more by the organization (i.e.,
organizational structure or work assignment). Some external cues (i.e., in-
terdependence, task difficulty) will influence self-efficacy only to the extent
that they are recognized in advance by the individual. Also, some of the
internal factors, like effort, can be influenced by organizational interven-
tions (e.g., bonuses for hard work), and people can be taught how to im-
prove their control over the environment through training in assertiveness,
influence tactics, impression management, leadership skills, and so on.
Self-efficacy level is limited in a variety of ways by the controllability
of the determinants. Some determinants are not controllable by anyone

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
202 Academy of Management Review April

(e.g., weather). Other determinants may be controllable for long-term de-

velopment but not for immediate change (e.g., personality, physical fitness,
extent of knowledge and skills). Still other determinants are controllable
only to the extent that they are recognized, or acted upon by the organiza-
tion or the individual (e.g., task complexity, appropriate strategies). An
understanding of both the task and the individual is needed if self-efficacy
is to be enhanced.
The Process of Change
Recall that self-efficacy mediates the translation of knowledge and abil-
ities into skilled performance (Bandura, 1988b). Two potential contributors to
this process deserve further attention: the attributional process and feed-
back.
Attributions about the causes of performance differ for people with high
and low self-efficacy. For example, Collins (1982) and Silver and his col-
leagues (1991) demonstrated that people with high and low self-efficacy
interpret performance feedback similarly when successful, but dissimilarly
when unsuccessful. When a person with high self-efficacy and a person
with low self-efficacy both succeed (get a problem right), they attribute that
success to the presence of ability. Under conditions of failure, people with
high self-efficacy attribute their failure to insufficient effort or bad luck,
whereas people with low self-efficacy attribute their failure to lack of ability.
Because ability may be viewed as a relatively stable determinant of per-
formance, self-efficacy and effort may decline with subsequent attempts at
the task. In such a case, a spiraling downward of performance, called an
exacerbation cycle (Storms & McCaul, 1976), may occur, and this may be
difficult to reverse.
One major way to change self-efficacy is to intervene in this attribu-
tional process (Fosterling, 1985). This intervention is especially important if
there is reason to believe that causal attributions are inaccurate, especially
attributions about stable abilities or traits. The adoption of new beliefs about
performance capability can be facilitated by the provision of accurate in-
formation about the causes of performance, as well as information about the
specific tasks that the employee is doing well or poorly.
Thus, when attributional errors are made, it is clear that frequent and
detailed feedback should enhance the accuracy of self-efficacy. However,
the mechanism by which it does so needs to be understood. Feedback
clarifies person-performance contingencies that may be used in the revision
of self-efficacy. Bandura (1986, 1988b) conceptualized these contingencies in
terms of an image-matching process. Individuals have a cognitive repre-
sentation of efficacious action based on past experience, learning, obser-
vations, and the integration and synthesis of this knowledge (cf. Figure 1).
These images guide behavior and function as internal standards for
change. Feedback on task performance provides information for detecting
as well as correcting discrepancies between images, actions, and out-
comes. When performance is substandard, an accurate conception of effi-

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 203

cacy determinants will facilitate an employee's readiness to receive appro-

priate interventions. For example, an employee who inaccurately attributes
performance failures to lack of effort may be reluctant to acknowledge the
need for more skill training. However, frequent, specific feedback may help
the employee correct improper attributions and participate actively in train-
ing activities.
Traditional approaches to improving performance have encompassed
task-relevant improvements (i.e., work design and working conditions), mo-
tivational techniques (goal setting, reinforcement, etc.), and improvements
in ability through training. Here, it has been discussed how self-efficacy can
mediate the effects of these changes on performance. Evidence also has
demonstrated that self-efficacy influences the degree of skill acquisition and
retention in learning situations (Gist et al., 1989; Gist, Stevens, & Bavetta,
1991; Wood & Locke, 1987). Efforts to improve employee abilities (through
classroom or on-the-job training) can be facilitated by understanding how to
improve self-efficacy, and the ideas developed thus far suggest several
strategies for improving self-efficacy and performance.
Intervention Strategies
A review of Figure 2 suggests that when the individual is engaged in
efficacy analysis, major questions are asked: what is required by the task,
how much can I offer under the situation, and what is the relative contri-
bution of each performance determinant. From this perspective, three strat-
egies for changing self-efficacy emerge:
Strategy 1: Provide information that gives the individual a
more thorough understanding of the task attributes, com-
plexity, task environment (primarily through the use of
mastery and modeling experiences), and the way in
which these factors can be best controlled.
Strategy 2: Provide training that directly improves the in-
dividual's abilities or understanding of how to use abili-
ties successfully in performing the task (primarily through
the use of mastery, modeling, and persuasion experi-
ences).
Strategy 3: Provide information that improves the individ-
ual's understanding of behavioral, analytical, or psycho-
logical performance strategies or effort expenditure re-
quired for task performance (primarily through the use of
modeling, feedback and persuasion).
Individually, or in combination, these methods should lead to changes
in self-efficacy accuracy and level. Strategies 1 and 2 entail the use of some
fairly standard procedures, including on-the-job training, work simulations
and samples, assessment techniques and centers, mentoring, counseling,
job rotation, and apprenticeships. However, because self-efficacy is partly
a motivational construct, the importance of individual determinants must be

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
204 Academy of Management Review April

weighed heavily in any consideration of how to alter self-efficacy and per-

formance. Strategy 3 is aimed at the highly variable, internal determinants,
which are generally the most individually controllable and may influence
self-efficacy most immediately; thus, it deserves particular attention and it
can be utilized in the following ways.
First, persuasion should take the form of counseling or coaching to
clarify the pros and cons of various performance strategies. Obviously,
counseling can specify the behavioral and analytical strategies that lead to
better performance. However, counseling also can address psychological
strategies that may increase task performance-particularly the way in
which these strategies may be used to counteract task environmental con-
cerns (e.g., imaging or relaxation).
Second, training may be required to improve an awareness of correct
strategies or effort considerations. The role of behavioral modeling is well
established in the training literature, yet most modeling interventions focus
primarily on behavioral performance strategies (i.e., an observable se-
quence of actions). Successful performance may require the effective use of
other types of performance strategies. Positive results have been found
through the use of cognitive modeling of psychological performance strat-
egies (i.e., the use of self-statements to mitigate inhibitions and guide per-
formance) during training for idea generation among managers (Gist,
1989). The effectiveness of this method of training should be explored for
other tasks in which psychological performance strategies may be impor-
tant and for tasks involving analytical performance strategies (such as cor-
rect judgment factors and sequences in decision making).
A final utilization of Strategy 3 is suggested in a study by Bandura and
Schunk (1981), who found that task interest increased, along with perfor-
mance and self-efficacy, when incremental subgoals were used. Thus,
training in proximal goal setting may facilitate performance (by increasing
effort and interest) in cases where employees have low self-efficacy.
Each of the three major strategies for changing self-efficacy rests on the
provision of information. However, even though feedback must play an
important role in the revision of self-efficacy, the content of this information
may vary. For example, some evidence exists that distorted information
may be useful for increasing self-efficacy. Studies have shown that video
feedback of prior performance can be edited, as a form of "self"-modeling,
to enhance performance (e.g., Gonzales & Dowrick, 1982). One way of
doing this is to remove (edit out) the individual's mistakes (selective infor-
mation); another is to edit the mistakes themselves to show correct perfor-
mance (positive misinformation). Apparently, the first method alters the in-
dividual's assessment of task requirements and ability level (Strategies 1
and 2). The second method might provide information (feedforward) on
more effective task strategies (Strategy 3). Collectively, these data can be
used to revise self-efficacy. Findings in the area of selective and distorted
feedback are supported by those of Eden and Ravid (1982), who studied the
effects of self-expectancies. They found that the mere induction of positive

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 205

self-expectancies (through verbal feedback) initiated and sustained higher

performance for one group of trainees when compared to a control group
not given the positive feedback.
The use of distorted information has ethical and practical implications
that dictate caution. When individuals do not have the ability to perform
well on a particular task, continued exhortations to work harder are likely to
increase, rather than decrease, the exacerbation cycle and may be detri-
mental to self-esteem over time. The assumption that permanent, positive
changes in an individual's self-efficacy can be established for every task
may be both incorrect and harmful. In employment settings, attention
should be focused on training methods that enhance both the motivation to
learn and skill mastery, as well as self-efficacy.

SUMMARYAND IMPLICATIONS
Bandura has argued convincingly that "human accomplishments and
positive well being require an optimistic and resilient sense of personal
efficacy" (1988a: 49). The questions addressed here pertain to how these
judgments of efficacy are made and how they can be changed. The process
involved in forming self-efficacy (Figure 1) was conceptualized, the cues
that people use in constructing their beliefs (Figure 2) were described and
categorized, and theoretical issues related to enhancing self-efficacy were
examined (Figure 3).
Much of Bandura's work has focused on how self-efficacy influences
subsequent thought and motivational mechanisms, which in turn influence
performance. Although current work with colleagues such as Cervone,
Schunk, and Wood has focused on education or decision-making processes,
some of the earlier developmental work on this theory was conducted in
clinical settings with phobias. Thus, several implications exist for the exten-
sion of self-efficacy theory and practice in organizational settings.
First, further research is needed on the formation of self-efficacy. Even
though this paper conceptualizes a number of self-efficacy determinants,
processes by which meaning is inferred from information cues, and a per-
spective on how the estimation of orchestration capacity occurs, empirical
validation of these is needed. Attributional analyses of past experience
have been studied widely; however, the mechanisms by which individuals
analyze task requirements and assess personal and situational resources or
constraints in new or revised task situations are less clear. Therefore, the
following propositions are offered.
Proposition 1: In novel task situations, self-efficacy is
formed through an assessment of task requirements and
personal and situational resources and constraints. The
identification and weighting of multiple cues in estimat-
ing orchestration capacity (self-efficacy) in these situa-
tions is subject to the constraints and processes involved
in the formation of other complex judgments.

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
206 Academy of Management Review April

Proposition 2: As task experience increases, individuals

rely more heavily on simple attributional analysis of past
performance and current motivation as determinants of
self-efficacy rather than on more comprehensive assess-
ments of task requirements and personal and situational
resources and constraints.
Next, the measurement of self-efficacy may be improved through atten-
tion to task complexity. Some evidence suggests that the predictive validity
of self-efficacy for performance on complex tasks may be weaker than for
performance on simple tasks (cf. Stumpf et al., 1987; Taylor et al., 1984;
Wood & Locke, 1987), probably reflecting lower accuracy in assessing task
requirements and individual or situational resources or constraints for these
tasks. When self-efficacy is measured, attention to the subcomponents or
subordinate processes required for the performance of complex tasks may
enhance validity. However, some subordinate processes (e.g., creative
thought) may not lend themselves well to the delineation of specific difficulty
levels used in traditional measures of self-efficacy. Thus, further psycho-
metric studies of self-efficacy for complex tasks are needed. Valid process
measures could be used as a diagnostic tool for developmental needs (e.g.,
training and self-efficacy change), as well as for predicting performance on
complex tasks with greater accuracy. Therefore,
Proposition 3: The predictive validity of self-efficacy is
dependent not only on the measurement protocol but on
the complexity of the task and performance criteria.
Also, more empirical studies in work-related settings are needed to
develop targeted interventions for enhancing self-efficacy. Training tech-
niques including actual experience, modeling, and persuasion can be ef-
fective. Yet more knowledge is needed about which self-efficacy determi-
nants are influenced most by which methods and the boundary conditions
on the effectiveness of various interventions. Thus,
Proposition 4: The effectiveness of interventions designed
to increase self-efficacy will depend on the determinants
(e.g., perceptions of abilities, task complexity) that are
affected by the interventions and on the weighted contri-
bution of these determinants to performance.
Because changing self-efficacy may require changing the way in
which individuals process information and make attributions, studies are
needed on the value of alternative forms of feedback for individuals who
have low and high self-efficacy. Specific feedback regarding the subordi-
nate processes involved in performance (e.g., strategies) may be more valu-
able than outcome feedback for enhancing self-efficacy on some tasks.
Also, studies are needed to determine the effects that negative feedback has
on the performance of subjects with low self-efficacy. Negative feedback
may exacerbate an already fragile image of person-performance contin-
gencies, leading to lower subsequent performance. Operant conditioning
suggests that positive feedback (praise) on process components that may

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 207

have been performed successfully, coupled with the establishment of incre-

mental subgoals on weaker process components, might be ideal for break-
ing exacerbation cycles among individuals with low self-efficacy. There-
fore,
Proposition 5: Frequent feedback can increase the accu-
racy of self-efficacy as well as motivation. To be maxi-
mally effective, the content and sign of the feedback must
be appropriate for the individual (based on self-efficacy
level and the task).
The significance of self-efficacy for motivation and performance in work
settings has been well demonstrated, although many questions remain
about how efficacy perceptions are formed and how they influence behav-
ioral choices. This article provides a theoretical analysis of the determinants
of self-efficacy and its malleability by integrating ideas from several do-
mains. A number of implications of this analysis exist for research and
practice in organizational behavior. Research should be pursued regarding
the theoretical issues identified, the methods and techniques to enhance
self-efficacy, and the understanding of how and why these interventions
work. Knowledge gained from addressing these issues would be valuable
not only to researchers but also to practitioners concerned with improving
employee performance.
REFERENCES
Ajzen, I., & Fishbein, M. 1980. Understanding attitudes and predicting social behavior. En-
glewood Cliffs, NJ: Prentice-Hall.
Anderson, C. A. 1983. Motivational and performance deficits in interpersonal settings: The
effects of attributional style. Journal of Personality and Social Psychology, 45: 1136-1147.
Anderson, N. H., & Zalinski, J. 1988. Functional measurement approach to self estimation in
multiattribute evaluation. Journal of Behavioral Decision Making, 1: 191-221.
Andrews, V. R., & Debus, R. L. 1978. Persistence and the causal perception of failure: Modi-
fying cognitive attributions. Journal of Educational Psychology, 70: 151-166.
Bandura, A. 1977. Social learning theory. Englewood Cliffs, NJ:Prentice-Hall.
Bandura, A. 1982. Self-efficacy mechanism in human agency. American Psychologist, 37:
122-147.
Bandura, A. 1984. Recycling misconceptions of perceived self-efficacy. Cognitive Therapy and
Research, 8: 213-229.
Bandura A. 1986. Social foundations of thought and action: A social-cognitive view. Engle-
wood Cliffs, NJ: Prentice-Hall.
Bandura, A. 1988a. Self-regulation of motivation and action through goal systems. In V. Ham-
ilton, G. H. Bower, & N. H. Frijda (Eds.), Cognitive perspectives on emotion and motiva-
tion: 37-61. Dordrecht, Netherlands: Kluwer Academic Publishers.
Bandura, A. 1988b. Reflection on nonability determinants of competence. In R. J. Sternberg &
J. Kolligian, Jr. (Eds.), Competence considered: Perceptions of competence and incom-
petence across the lifespan: 315-362. New Haven, CT: Yale University Press.
Bandura, A. 1988c. Organizational applications of social cognitive theory. Australian Journal
of Management, 13: 137-164.
Bandura, A., & Cervone, D. 1986. Differential engagement of self-reactive mechanisms gov-

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
208 Academy of Management Review April

erning the motivational effects of goal systems. Organizational Behavior and Human
Decision Processes, 38: 92-113.
Bandura, A., & Schunk, D. 1981. Cultivating competence, self-efficacy, and intrinsic interest
through proximal self-motivation. Journal of Personality and Social Psychology, 41: 586-
598.
Bandura, A., & Wood, R. E. 1989. Effect of perceived controllability and performance stan-
dards on self-regulation of complex decision making. Journal of Personality and Social
Psychology, 56: 805-814.
Barling, J., & Beattie, R. 1983. Self-efficacy beliefs and sales performance. Journal of Organi-
zational Behavior Management, 5: 41 -51.
Beach, L. R. 1967. Multiple regression as a model for human information utilization. Organi-
zational Behavior and Human Performance, 2: 276-289.
Beach, L. R., & Mitchell, T. R. 1978. A contingency model for the selection of decision strate-
gies. Academy of Management Review, 3: 439-449.
Brockner, J. 1979. The effects of self-esteem, success-failure, and self-consciousness on task
performance. Journal of Personality and Social Psychology, 37: 1732-1741.
Brockner, J. 1988. Self-esteem at work. Lexington, MA: Lexington Books.
Brunswick, E. 1956. Perception and the representative design of psychological experiments.
Berkeley: University of California Press.
Campbell, D. J. 1988. Task complexity: A review and analysis. Academy of Management
Review, 13: 40-52.
Campbell, D. T., & Fiske, D. W. 1959. Convergent and discriminant validation by the multi-
trait-multimethod matrix. Psychological Bulletin, 56: 81-105.
Campbell, N. K., & Hackett, G. 1986. The effects of mathematics task performance on math
self-efficacy and task interest. Journal of Vocational Behavior, 28: 149-162.
Cervone, D. 1985. Self-efficacy judgment under uncertainty: Availability biases in perceived
self-efficacy and behavior. Unpublished doctoral dissertation, Stanford University, Stan-
ford, CA.
Cervone, D., & Peake, P. K. 1986. Anchoring, efficacy, and action: The influence of judgmental
heuristics on self-efficacy judgments and behavior. Journal of Personality and Social
Psychology, 50: 492-501.
Chaiken, S. 1986. The heuristic model of persuasion. In M. P. Zanna, J. M. Olson, & C. P.
Herman (Eds.), Social influence: The Ontario symposium, vol. 5: 89-134. Hillsdale, NJ:
Erlbaum.
Collins, J. L. 1982. Self-efficacy and ability in achievement behavior. Paper presented at the
annual meeting of the American Educational Research Association, New York.
Cook, T. D., & Campbell, D. T. 1979. Quasi-experimentation: Design and analysis issues for
field settings. Boston: Houghton Mifflin.
Decker, P. J. 1980. Effects of symbolic coding and rehearsal in behavior-modeling training.
Journal of Applied Psychology, 65: 627-634.
Earley, P. C., Connelly, T., & Ekegren, G. 1989. Goals, strategy development, and task per-
formance: Some limits on the efficacy of goal setting. Journal of Applied Psychology, 74:
24-33.
Eastman, C., & Marzillier, J. S. 1984. Theoretical and methodological difficulties in Bandura's
self-efficacy theory. Cognitive Therapy and Research, 8: 213-229.
Eden, D., & Ravid, G. 1982. Pygmalion versus self-efficacy: Effects of instructor- and self-
expectancy on trainee performance. Organizational Behavior and Human Performance,
30: 351-364.

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 209

Einhorn, H. J., & Hogarth, R. M. 1981. Behavioral decision theory: Processes of judgment and
choice. Annual Review of Psychology, 32: 53-88.
Fazio, R. H. 1986. How do attitudes guide behavior? In R. M. Sorrentino & E. T. Higgins (Eds.),
Handbook of motivation and cognition: 204-243. New York: Guilford.
Feldman, J. M. 1981. Beyond attribution theory: Cognitive processes in performance evalua-
tion. journal of Applied Psychology, 66: 127-148.
Fishbein, M., & Ajzen, I. 1975. Belief, attitude, intention, and behavior. Reading, MA: Add-
ison-Wesley.
Friedman, M., & Rosenman, R. 1974. Type A behavior and your heart. New York: Knopf.
Fosterling, F. 1985. Attributional retraining: A review. Psychological Bulletin, 98: 495-512.
Frayne, C. A., & Latham, G. P. 1987. Application of social learning theory to employee self-
management of attendance. Journal of Applied Psychology, 72: 387-392.
Gersick, C. J. G., & Hackman, J. R. 1990. Habitual routines in task performing groups. Orga-
nizational Behavior and Human Decision Processes, 47: 65-97.
Gist, M. E. 1987. Self-efficacy: Implications for organizational behavior and human resource
management. Academy of Management Review, 12: 472-485.
Gist, M. E. 1989. The influence of training method on self-efficacy and idea generation among
managers. Personal Psychology, 42: 787-805.
Gist, M. E., Schwoerer, C., & Rosen, B. 1989. Effects of alternative training methods on self-
efficacy and performance in computer software training. Journal of Applied Psychology,
74: 884-891.
Gist, M. E., Stevens, C. K., & Bavetta, A. B. 1991. The influence of self-efficacy and training
condition on retention of learning. Personnel Psychology, 44: 837-861.
Gonzales, F., & Dowrick, P. 1982. Mechanisms of self-modeling: An investigation of skills
acquisition versus self-belief. Unpublished manuscript, University of Alaska, Anchorage.
Graik, F. I. M., & Lockhart, R. S. 1972. Level of processing: A framework for memory research.
Journal of Verbal Learning and Nonverbal Behavior, 11: 671-684.
Hill, T., Smith, N. D., & Mann, M. F. 1987. Role of efficacy expectations in predicting the
decision to use advanced technologies. Journal of Applied Psychology, 72: 307-314.
Ilgen, D. R., Nebeker, D. M., & Pritchard, R. D. 1981. Expectancy theory measures: An empir-
ical comparison in an experimental simulation. Organizational Behavior and Human
Performance, 28: 189-223.
Kanfer, R. 1991. Motivation theory and industrial/organizational psychology. In M. D. Dunnette
(Ed.), Handbook of industrial and organizational psychology, vol. 1. Palo Alto, CA: Con-
sulting Psychologists Press.
Kanfer, R., & Ackerman, P. L. 1989. Motivation and cognitive abilities: An integrative/aptitude-
treatment interaction approach to skill acquisition. Journal of Applied Psychology, 74:
657-689.
Kavanagh, D. J., & Bower, G. H. 1985. Mood and self-efficacy: Impact of joy and sadness on
perceived capabilities. Cognitive Therapy and Research, 9: 507-525.
Kelley, H. H. 1971. Attributions in social interactions. Morristown, NJ:General Learning Press.
Kelley, H. H., & Michela, J. L. 1980. Attribution theory and research. Annual Review of Psy-
chology, 31: 457-501.
Kiggundu, M. N. 1981. Task interdependence and the theory of job design. Academy of Man-
agement Review, 6: 499-508.

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
210 Academy of Management Review April

Klein, H. J. 1989. An integrated control theory model of work motivation. Academy of Man-
agement Review, 14: 150-172.
Landy, F. J. 1986. Stamp collecting versus science: Validation as hypothesis testing. American
Psychologist, 41: 1183-1192.
Langer, E. G. 1989. Minding matters: The mindlessness; mindfulness theory of cognitive activ-
ity. In L. Berkowitz (Ed.), Advances in experimental social psychology, 22: 137-173. New
York: Academic Press.
Lazarus, R. S., & Folkman, S. 1984. Stress, appraisal, and coping. New York: Springer.
Lent, R. W., Brown, S. D., & Larkin, K. C. 1987. Comparison of three theoretically derived
variables in predicting career and academic behavior: Self-efficacy, interest congruence,
and consequence thinking. Journal of Counseling Psychology, 34: 293-298.
Locke, E. A., & Latham, G. P. 1990. A theory of goal setting and task performance. Englewood
Cliffs, NJ: Prentice-Hall.
Locke, E. A., Frederick, E., Lee, C., & Bobko, P. 1984. Effect of self-efficacy, goals, and task
strategies on task performance. Journal of Applied Psychology, 69: 241-251.
Mikulincer, M., & Nizan, B. 1988. Causal attribution, cognitive interference, and the general-
ization of learned helplessness. Journal of Personality and Social Psychology, 55: 470-
478.
Mitchell, T. R. 1974. Expectancy models of satisfaction, occupational preference, and effort: A
theoretical, methodological, and empirical appraisal. Psychological Bulletin, 81: 1053-
1077.
Mitchell, T. R., & Beach, L. R. 1990. Do I love thee? Let me count: Towards an understanding
of intuitive and automatic decision making. Organizational Behavior and Human Deci-
sion Processes, 47: 1-20.
Norman, D. A., & Bobrow, D. B. 1975. On data limited and resource limited processes. Cog-
nitive Psychology, 7: 44-64.
Petty, R. E., & Cacioppo, J. T. 1986. The elaboration likelihood model of persuasion. In L.
Berkowitz (Ed.), Advances in experimental social psychology, vol. 19: 87-136. New York:
Academic Press.
Russell, D., & McAuley, E. 1986. Causal attributions, causal dimensions, and affective reac-
tions to success and failure. Journal of Personality and Social Psychology, 50: 154-166.
Schneider, W., & Schriffin, R. M. 1977. Controlled and automatic human information process-
ing: Controlled detection, search, and attention. Psychological Review, 84: 1-66.
Schunk, D. H. 1983. Ability versus effort attributional feedback: Differential effects on self-
efficacy and achievement. Journal of Educational Psychology, 75: 848-856.
Schunk, D. H. 1984. Sequential attributional feedback and children's achievement behavior.
Journal of Educational Psychology, 76: 1159-1169.
Schunk, D. H., & Gunn, R. P. 1986. Self-efficacy and skill development: Influence of task strat-
egies and attributions. Journal of Educational Research, 79: 238-244.
Seligman, M. E. P. 1975. Helplessness: On depression, development, and death. San Fran-
cisco: Freeman.
Shrauger, J. S. 1972. Self-esteem and reactions to being observed by others. Journal of Per-
sonality and Social Psychology, 23: 192-200.
Silver, W. S., Mitchell, T. R., & Gist, M. E. 1991. Interpreting performance information: The
influence of self-efficacy on causal attributions for successful and unsuccessful perfor-
mance. Unpublished manuscript, University of Washington, Seattle.
Slovic, P., & Lichtenstein, S. 1971. Comparison of Bayesian and regression approaches to the

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions
1992 Gist and Mitchell 211

study of information processing in judgment. Organizational Behavior and Human Per-

formance, 6: 649-744.
Storms, M. D., & McCaul, K. D. 1976. Attribution processes and emotional exacerbation of
dysfunctional behavior. In J. Harvey, W. Ickes, & R. F. Kidd (Eds.), New directions in
attribution research, vol. 1: 143-164. Hillsdale, NJ: Erlbaum.
Stumpf, S. A., Brief, A. P., & Hartman, K. 1987. Self-efficacy expectations and coping with
career-related events. Journal of Vocational Behavior, 31: 91-108.
Taylor, M. S., Locke, E. A., Lee, C., & Gist, M. E. 1984. Type A behavior and faculty research
productivity: What are the mechanisms? Organizational Behavior and Human Decision
Processes, 34: 402-418.
Taylor, S. E., & Crocker, H. G. 1981. Schematic bases of social information processing. In E. T.
Higgins, C. P. Herman, M. P. Zanna (Eds.), Social cognition: The Ontario symposium, 1:
89-134. Hillsdale, NJ: Erlbaum.
Vroom, V. H. 1964. Work and motivation. New York: Wiley.
Weiner, B. 1979. A theory of motivation for some classroom experiences. Journal of Educa-
tional Psychology, 71: 3-25.
Weiner, B. 1985. An attributional theory of achievement, motivation and emotion. Psycholog-
ical Review, 92: 548-573.
Weiss, H. M., & Ilgen, D. R. 1985. Routinized behavior in organizations. Journal of Behavioral
Economics, 14: 57-67.
Wickens, C. D. 1984. Processing resources in attention. In R. Parasuraman & D. R. Davies
(Eds.), Varieties of attention: 63-102. New York: Academic Press.
Wilson, T. D., & Linville, P. W. 1985. Improving the performance of college freshmen with
attributional techniques. Journal of Personality and Social Psychology, 49: 287-293.
Wood, R. E. 1986. Task complexity: Definition of the construct. Organizational Behavior and
Human Decision Processes, 37: 60-82.
Wood, R. E., & Bandura, A. 1989a. Impact of conceptions of ability on self-regulatory mecha-
nisms and complex decision making. Journal of Personality and Social Psychology, 56:
407-415.
Wood, R. E., & Bandura, A. 1989b. Social cognitive theory of organizational management.
Academy of Management Review, 14: 361-384.
Wood, R. E., & Locke, E. A. 1987. The relation of self-efficacy and grade goals to academic
performance. Educational and Psychological Measurement, 47: 1013-1024.
Zalesny, M. D., & Ford, J. K. 1990. Extending the social information processing perspective:
New links to attitudes, behaviors, and perceptions. Organizational Behavior and Human
Decision Processes, 47: 205-246.
Zoeller, L. J., Mahoney, G., & Weiner, B. 1983. Effects of attribution training on the assembly
task performance of mentally retarded adults. American Journal of Mental Deficiency, 88:
109-112.

Marilyn Gist received her Ph.D. from the University of Maryland. She is an associate
professor of organizational behavior and human resource management in the Grad-
uate School of Business Administration at the University of Washington. Her research
interests focus on self-efficacy as it relates to other social cognitive influences on mo-
tivation, training, and performance.
Terence Mitchell received his Ph.D. from the University of Illinois. He holds the Carlson
professorship in the Graduate School of Business Administration at the University of
Washington. His current research interests include the study of social cognitive influ-
ences on decision making, motivation, and human performance.

This content downloaded from 195.78.109.78 on Fri, 9 May 2014 14:28:03 PM

All use subject to JSTOR Terms and Conditions