Vous êtes sur la page 1sur 6

Journal of Applied Psychology Copyright 1994 by the American Psychological Association, Inc.

1994, Vol. 79. No. 3, 364-369 0021-90IO/94/S3.00

Self-Efficacy Beliefs: Comparison of Five Measures


Cynthia Lee and Philip Bobko

In 2 studies, the researchers compared 5 ways of operationalizing self-efficacy that are commonly
found in the literature and assessed the antecedents and consequences of self-efficacy on the basis of
A. Bandura's (1986) conceptualization. Results indicated that measuring self-efficacy by using a
task-specific, 1-item confidence rating showed the lowest convergent validity with the other self-
efficacy operationalizations and showed the least consistency in its correlation with the hypothesized
self-efficacy antecedents and outcomes. Furthermore, self-efficacy magnitude and self-efficacy
strength (combining all the certainty answers) appeared to be inferior to self-efficacy composites
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

based on combining only the strength items where the magnitude response was "yes, I can perform
This document is copyrighted by the American Psychological Association or one of its allied publishers.

at that level."

The construct of self-efficacy has received increasing atten- method of measuring self-efficacy is to ask subjects to rate their
tion in the literature (Gist & Mitchell, 1992). Bandura (1986) self-efficacy strength (e.g., Bandura & Jourden, 1991; Bandura
defined perceived self-efficacy as follows: & Wood, 1989; Brown, Lent, & Larkin, 1989; Matsui & Tsuka-
moto, 1991). (Note that, for each definition, our literature re-
people's judgments of their capabilities to organize and execute view uncovered many studies. We present only a few here be-
courses of action required to attain designated types of perfor-
mances. It is concerned not with the skills one has but with the cause of space considerations.) Self-efficacy level (or magnitude)
judgments of what one can do with whatever skills one possesses, is a second popular method for assessing self-efficacy (e.g., Cer-
(p. 391) vone & Palmer, 1990; Peake & Cervone, 1989). Also, quite a
number of studies have presented results on both the strength
Those who have a strong sense of self-efficacy in a particular and magnitude of self-efficacy (e.g., Berry, West, & Dennehey,
situation will devote their attention and effort to the demands 1989; Locke, Frederick, Lee, & Bobko, 1984; Matsui, Ikeda,
of the situation, and when faced with obstacles and difficult sit- & Ohnishi, 1989; Podsakoff & Farh, 1989; Rebok & Balcerak,
uations, these individuals will try harder and persist longer. Such 1989).
individuals are also inclined to attribute failures on difficult As a third possibility, some researchers have used a combina-
tasks to insufficient effort. tion of the magnitude and strength indexes. For example, Lee
When operationally measuring self-efficacy, researchers typi- (1988), Podsakoff and Farh (1989), and Taylor, Locke, Lee, and
cally ask individuals whether they can perform at specific levels Gist (1984) standardized the self-efficacy strength items by con-
on a specific task (responses are either yes or no) and ask for verting them to z scores and then summed across only those
the degree of confidence in that endorsement (rated on a near- self-efficacy levels to which subjects answered yes. A fourth
continuous scale from total uncertainty to total certainty) at method is a variant of the third (e.g., Frayne & Latham, 1987;
each specific performance level. One measure of self-efficacy, Gist, Schwoerer, & Rosen, 1989; McAuley, Wraith, & Duncan,
labeled self-efficacy magnitude, is formed by summing the total 1991), except that one uses the raw scores on self-efficacy
positive responses (i.e., the number of yes responses) given by a strength instead of adding the z scores. A fifth and final method
subject. Yet a different measure, self-efficacy strength, is formed of assessing self-efficacy strength is to use a single item for which
by summing the confidence ratings across all performance lev- respondents rate their confidence level on a given task (e.g.,
els. Many published empirical studies have used either self- Clement, 1987) or on each investment decision (Kerr, 1989).
efficacy magnitude or self-efficacy strength. There are many other ways to measure self-efficacy, such as the
Our review of the literature indicated that the most common use of multiple items with 14-point Likert-type scales (Sherer et
al., 1982), 7-point Likert-type scales (e.g., Barling & Beattie,
1983; Jones, 1986), 5-point Likert-type scales (Gould, Hodge,
Cynthia Lee, College of Business Administration, Northeastern Uni- Peterson, & Giannini, 1989), 4-point anchored scales (e.g.,
versity; Philip Bobko, Department of Management, Rutgers University. Bores-Rangel, Church, Szendre, & Reeves, 1990), or a 3-point
Preparation of this article was facilitated by the Riesman Award to forced-choice format (e.g., Holahan & Holahan, 1987). These
Cynthia Lee from the College of Business Administration at Northeast- measures do not correspond to Bandura's (1986) recommenda-
ern University. tion for assessing the strength and magnitude of self-efficacy
We thank Tom Lee, Edwin Locke, Karyll Shaw, and several anony- and, thus, are excluded from the present study.
mous reviewers for their comments. We also thank Sharon Beckstrom,
Although studies on self-efficacy demonstrate its importance
Joseph Czajka, Jeffrey Mello, Leticia Pena, and Frank Spital for their
in predicting and improving work performance, none of the
assistance in data collection.
Correspondence concerning this article should be addressed to Cyn- empirical studies focus on the correlations among the self-
thia Lee, College of Business Administration, Northeastern University, efficacy measures. The purpose of our investigation is to exam-
304 Hayden Hall, Boston, Massachusetts 02115. ine the convergent validity of the five self-efficacy operationali-
364
SELF-EFFICACY MEASURES 365

zations and to present a validation of a partial nomological net- Hypothesis 3: Consistent with prior literature, there will be positive
relationships between self-efficacy and both self-set goals and per-
work incorporating these self-efficacy operationalizations. We formance. More important, and consistent with earlier logic, the
conducted two studies, using two different tasks, with students composite measures of self-efficacy will show the highest validities,
as subjects. Study 1 was conducted in a natural classroom set- and the single-item measure will show the lowest validity, with
ting in which the subject matter (introduction to organizational these consequence measures.
behavior) served as the complex, but somewhat novel task.
Wood and Locke's (1987) academic self-efficacy scale (com-
posed of multiple task elements) was used to examine the con- Study 1
vergent and predictive validities of the self-efficacy magnitude, In Study 1, we investigated the convergent and predictive va-
strength, and composite measures. In Study 2 we assessed the lidities of the various operationalizations of self-efficacy level,
generalizability of Study 1 by using a simple and novel brain- strength, and composite operationalizations, using Wood and
storming task. Our self-efficacy measure was consistent with the Locke's (1987) academic self-efficacy measure.
one used by Locke et al. (1984), who also used the brainstorm-
ing task in their study. Thus, we arrived at the following
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Method
This document is copyrighted by the American Psychological Association or one of its allied publishers.

hypothesis.
Respondents and Procedures
Hypothesis 1: There will be positive relationships among all self-
efficacy operationalizations (some of this will be because two mea- Participants (N = 207) were predominantly third-year undergradu-
sures are components of the composite measures). More impor- ates in a 5-year degree program from six introductory management
tant, we expect that the single-item confidence rating will show the classes. There were 113 men and 83 women; 11 undergraduates did not
least convergence with the other four measures, because of the po- indicate their gender. The average age was 21 years. Students were asked
tential for generally low reliability, low validity, or both from single-
item ratings. to participate voluntarily in a study on personality. Individuals who
agreed to participate (approximately 90% of all the students) completed
Wood and Locke's (1987) academic self-efficacy questionnaire after the
students had taken their first examination in the course but before they
Antecedents of Self-Efficacy had received their grades. This timing was purposive, so that subjects
would have some information about their ability to cope with the course
Because we did not intend to review all the determinants of demands and yet would not base their ratings simply on their grades
self-efficacy in the literature (cf. Bandura, 1986), we only exam- (Wood & Locke, 1987).
ined past performance and affectivity as antecedents of self-
efficacy. Of the major sources of self-efficacy, both ability and Measures
past performance have consistently been found to be positively
related to self-efficacy (Bandura, 1986; Locke & Latham, 1990; Self-efficacy. Because our sample consisted of students, we used
Wood and Locke's (1987) academic self-efficacy scale.
Podsakoff& Farh, 1989; Wood & Bandura, 1989). Further-
As did Wood and Locke, we excluded items that had extremely
more, Kavanagh and Bower (1985) found that subjects in whom skewed distributions with very little or no dispersion. As a result, only
a happy mood had been induced reported higher overall self- two items out of five from each subscale were retained. For purposes of
efficacy than those in whom no mood had been induced. Ban- comparison with Wood and Locke's results, we also combined six of
dura (1986) further noted that if positive mood activates their seven subscales of class concentration, memorization, understand-
thoughts of accomplishments, perceived self-efficacy will be ing, explaining concepts, discriminating concepts, and note taking for
boosted. As noted in Study 2, we had the opportunity to collect subsequent analyses. The Cronbach coefficient alphas for each of these
data on positive mood states. This led to our second hypothesis. subscales were .86, .82, .80, .86, .81, and .84, respectively. The use of
coefficient alphas to estimate the reliability of self-efficacy strength and
Hypothesis 2: There will be a positive relationship between both magnitude is somewhat deceiving because Gunman's scaling was used
past performance and positive affectivity with self-efficacy (Study to construct the self-efficacy measures. We report coefficient alphas for
2 only). Consistent with Hypothesis 1, we expect the single-item the purposes of completeness because most empirical studies have used
measure to have the lowest validity with these antecedents. Fur- the same reliability estimates. In Study 2, the test-retest correlations
thermore, we suggest that the third and fourth measures of self- were .77 and .78 for self-efficacy strength and magnitude, respectively.
efficacy will have the highest validities with antecedent measures, Students were asked to indicate if they could achieve a certain level of
because of their information richness and because they reflect the attainment (yes or no) as well as their degree of confidence in their abil-
original conceptualization of the self-efficacy construct. ity to perform at that level (on a scale from 0 to 10). Self-efficacy level,
or magnitude, was defined as the total number of yess divided by the
total number of items.
We defined self-efficacy strength as the mean confidence rating, using
Consequences of Self-Efficacy
a scale ranging from completely unconfidenl (0) to completely confident
The overall thrust of self-efficacy research is to provide both a (10). We computed self-efficacy strength by summing all of the scores
mechanism that mediates behavior change and a parsimonious across items and then dividing by the total number of items.
Self-efficacy composite measures of strength and magnitude were
account of why and how different techniques effect change computed in two ways. In one method we took the raw scores of self-
(Bandura, 1986). Other consequences of self-efficacy include efficacy strength and then summed these across self-efficacy levels that
the setting of higher personal goals and higher task performance were answered yes (total self-efficacy). The other method was to stan-
(Bandura, 1986; Locke & Latham, 1990; Wood & Bandura, dardize the self-efficacy strength items by converting them to z scores
1989). With this in mind, we proposed our third hypothesis. and then sum them across all self-efficacy levels that were answered yes.
366 CYNTHIA LEE AND PHILIP BOBKO

The fifth operationalization of self-efficacy was a one-item confidence Table 2


rating obtained by asking respondents "How confident are you that you Zero-Order Correlations Between Self-Efficacy (SE)
will obtain the score that you are trying for in your next exam?"; sub- and Goals and Performance: Study 1
jects responded on a 5-point scale ranging from extremely confident (5)
to no confidence at all (1). Exam score Self-set goal
Personal goals (examination score). This one-item measure of the
next examination-score goal was adopted from Locke and Bryan's Measure r
xyi
(1968) earlier study of goals and academic performance. This item
measures the lowest score the individual would be satisfied with in the SE magnitude .06 .13* .23*
next examination. SE strength .15* .12* .49*
The first and second examination scores were used as measures of SE Composite 1 .21** .20** .48*
SE Composite 2 .25** .20** .53*
class performance. All of the examination scores were z transformed Confidence rating .32** .08 .25*
within each of the six classes.
Note. Values in the table are validity coefficients. The confidence rat-
Results ing consisted of one item. SE Composite 1 = raw scores of SE strength
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

summed across SE magnitude items that were answered yes; SE Com-


This document is copyrighted by the American Psychological Association or one of its allied publishers.

Table 1 shows the intercorrelations of the various operation- posite 2 = standardized SE strength items summed across SE magni-
alizations of self-efficacy. The mean correlation, after applying tude items that were answered yes. rxyi = correlation with first exami-
nation score; rx>a = correlation with second examination score; rfy3 =
Fisher's z transformation, was .62. (Note that the particularly correlation with second examination score's goal.
high correlations were somewhat due to common components */><.05. **p<.01.
across measures.) Consistent with Hypothesis 1, the one-item
confidence rating generally demonstrated the lowest convergent
validity with the remaining measures. One of the noncombina-
tion measures (strength) also demonstrated relatively weaker be tested to see if they are statistically different by using Hotel-
convergence. Table 2 shows the correlations of these operation- ling's test for dependent correlations (cf. Cohen & Cohen,
alizations of self-efficacy with examination scores and self-set 1983). Each test statistic is somewhat unique because of the
goals. The results supported Hypothesis 3; that is, although the changing magnitudes of the intercorrelations. However, given
self-efficacy operationalizations were all positively related to our sample sizes, differences in correlations over. 15 tend to be
self-set goals and task performance, the strongest and most con- statistically significant. Thus, for example, both of the compos-
sistent positive relationships tended to be from the two compos- ite measures had higher correlations with self-set goals than did
ite measures of self-efficacy. the single-item confidence rating of self-efficacy. The complete
set of statistical tests is available from Cynthia Lee.
As noted by Wood and Locke (1987), performance feedback
Summary
is usually task specific in laboratory studies. In contrast, an ex-
The results provided support for the convergent validity of amination score is feedback from multiple tasks (e.g., studying,
the self-efficacy measures. The one-item confidence rating and class attendance, note taking, and memorizing). This perfor-
self-efficacy strength appeared to have the lowest correlation mance feedback is only given for overall examination perfor-
with all other self-efficacy measures and tended to be relatively mance, not for the individual task components that bring the
weakly related to examination score and examination-score examination score about. Thus, the self-efficacy ratings for such
goal. With the exception of the one-item confidence rating and tasks are probably not as accurate as they would be in single-
self-efficacy magnitude, Study 1 shows that the two self-efficacy task studies. Furthermore, Gist and Mitchell (1992) stated that
composites and self-efficacy strength operationalizations corre- in complex task situations, the complexity of requiring individ-
lated highest with self-set goals. In interpreting the correlations uals to estimate numerous skill and motivational parameters
in Table 2, note that any two correlations within a column can may increase the error of assessment. Therefore, in Study 2 we
used a single, but novel, brainstorming task to further investi-
gate the convergent validity of the various operationalizations
of self-efficacy magnitude, strength, and composite measures.
Table 1 In addition, we again examined the relative validities of self-
Correlations Among Self-Efficacy (SE) Measures: Study 1 efficacy operationalizations to the hypothesized antecedents
Measure 1 and outcomes.

1. SE magnitude _
2. SE strength .54** Study 2
3. SE Composite 1 .94** .71**
4. SE Composite 2 .90** .28** .86** Method
.28**
.28**
5. Confidence rating .29** .13*
Respondents
Note. Ns = 163-205 for survey after the first exam. The confidence The subjects were 92 undergraduates from three introductory man-
rating consisted of one item. SE Composite 1 = raw scores of SE agement classes of a 5-year degree program. As part of another study
strength summed across SE magnitude items that were answered yes;
SE Composite 2 = standardized SE strength items summed across SE (Lee & Bobko, 1992), each class was told that they would be experienc-
magnitude items that were answered yes. ing a different form of learning (through the use of a brainstorming
*p<.05. **p<.0l. exercise). Although students in each class were asked to volunteer, they
SELF-EFFICACY MEASURES 367

all participated. After the exercise was over, students were debriefed and Table 3
thanked. The mean age of these students was 22 years. There were 33 Correlations Among Self-Efficacy (SE) Measures
women and 57 men; 2 students did not indicate their gender. After Two Practice Trials: Study 2
Measure 3
Task _
1. SE magnitude
In the brainstorming task subjects were asked to generate uses for 2. SE strength .59*
common objects, such as a brick, books, clothes hanger, paper clip, or
3. SE Composite 1 .76* .90*
cushion (cf. Locke et al., 1984). Two 3-min experimental trials followed 4. SE Composite 2 .73* .50* .85*
two initial practice trials. Different objects were used for each trial. 5. Confidence rating .31* .28* .40* .41*

Note. N = 92. SE Composite 1 = raw scores of SE strength summed


Conditions across SE magnitude items that were answered yes; SE Composite 2 =
standardized SE strength items summed across SE magnitude items
There were three assigned goal levels: (a) the somewhat difficult goal that were answered yes.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

(n = 35) of generating at least 8 uses for a single object in 3 min, (b) a


This document is copyrighted by the American Psychological Association or one of its allied publishers.

moderately difficult goal (n = 28) of generating 12 uses in 3 min, and (c)


a very difficult goal (n = 29) of generating 16 uses in 3 min. These goal
levels were achieved by 50%, 12%, and 5%, respectively, of the subjects
during a pilot study. Results
Table 3 shows the correlations among the self-efficacy mea-
Procedure sures in Study 2. The correlations are consistent with those of
Study 1. The mean correlation of the self-efficacy measures, us-
At the beginning of the class period, the task was explained and the ing Fisher's z transformation, was also .62. Thus, as expected,
subjects were then allowed two practice trials. After these practice trials, the results supported Hypothesis 1 regarding the convergent va-
subjects were given their assigned goals. Measures of perceived goal lidity of the self-efficacy measures. As hypothesized, the single-
difficulty and a self-efficacy scale were then administered. Subjects were
then asked for their personal goal for the next trial. Finally, subjects were item confidence rating demonstrated the lowest convergent va-
asked to brainstorm for two experimental trials. lidity with the remaining measures.
Part of Hypothesis 2 predicted positive relationships between
self-efficacy and the antecedent measures of past performance
Measures and positive affectivity. Our data (see the positive affectivity and
We measured individual task performance by the total number of past performance columns in Table 4) supported this hypothe-
uses given, deleting responses that were inappropriate (e.g., "break it"
for the object "brick") or duplications within the same trial.
We assessed a subject's self-set, or personal, goal by their responses to Table 4
a single item asking how many uses the subject intended to list for the Correlations of Self-Efficacy (SE) Measures With Antecedents
two practice trials and the two experimental trials.
and Outcome Measures
Self-efficacy was assessed in a manner consistent with Study 1. Sub-
jects were asked to indicate yes or no as to whether they could perform Goal Goal
at five different levels of proficiency (i.e., listing 4, 8, 12, 16, and 20 Measure PA 1 2 Past Present
uses). Proficiency level (self-efficacy magnitude) was measured by the
total number of yes responses for listing the uses for each object divided PA .37** .36** .37** .34**
by the number of items to which subjects responded with a yes. For each
of the five proficiency levels, subjects also estimated their confidence After two practice trials
(self-efficacy strength) about their present capability to perform at that SE magnitude .29** .58** .51** .52** .53**
level on a scale from no confidence at all (0) to totally confident (100). SE strength .22* .52** .50** .54** .51**
Inspection of the individual items (five levels of proficiency) indicated SE Composite 1 .28** .63** .63** .61** .65**
that two of the items had very little dispersion and were making negligi- SE Composite 2 .19* .53** .58** .53** .60**
ble contributions to the overall measure. As in Wood and Locke (1987), Confidence rating .41** .34** .41** .54** .48**
these items (of listing 4 and 8 uses of an object) were deleted from sub-
sequent analyses. The self-efficacy magnitude, strength, and composite Controlling for assigned goal
scores were computed in the same manner as in Study 1. Likewise, we SE magnitude .31** .55** .49** .53** .51**
again used a one-item confidence rating by asking respondents "How SE strength .23** .51** .49** .54** .50**
confident are you that you will attain the goal of listing at least 'X' uses SE Composite 1 .30** .60** .61** .63** .64**
in 3 minutes for the next trial?" The Cronbach alphas for the self-effi- SE Composite 2 .19* .53** .41** .53** .60**
cacy magnitude and strength measures obtained after the first two prac- Confidence rating .41** .43** .49** .57** .54**
tice trials were .73 and .79, respectively, and their respective alphas after
the first experimental trial were .74 and .77. Note. PA = positive affectivity; Goal 1 = self-set goal after the first two
We measured positive affectivity on a 10-item scale developed and practice trials; Goal 2 = self-set goal after the first experimental trial;
validated by Watson, Clark, and Tellegen (1988). Respondents indicated Past = past performance; Present = present performance; SE Compos-
ite 1 = raw scores of SE strength summed across SE magnitude items
on a 5-point Likert-type scale the extent to which they felt, for example, that were answered yes; SE Composite 2 = standardized SE strength
"excited," "enthusiastic," "determined," and "attentive" over the past items summed across SE magnitude items that were answered yes.
year. The Cronbach alpha for this scale was .88. *p<.05. **p<.Q[.
368 CYNTHIA LEE AND PHILIP BOBKO

sis for all of the self-efficacy operationalizations. However, these sure by using a past-tense orientation. So both positive affec-
correlations were about equal when the antecedent measure was tivity and past performance differ from the goal measures and
past performance and the confidence rating was a higher corre- present performance, with the latter having a forward-looking
late of affectivity than the other self-efficacy measures. Thus, orientation. Bandura (1986) maintained that "people see out-
the differential predictions in Hypothesis 2 were not supported comes as contingent on the adequacy of their performances, and
in Study 2. care about those outcomes, that they rely on self-judged efficacy
In regard to Hypothesis 3, the correlations shown in Table 4 in deciding which courses of action to pursue" (p. 392). This
strongly supported the validity of the measures. As expected, distinction between past- and future-oriented measures may, in
the self-efficacy operationalizations were positively associated part, explain the unexpected large correlations of the confi-
with the outcomes of self-set goals and task performance. More dence rating with the positive affectivity and past performance
important, however, is that the two composite measures of self- measures.
efficacy were generally the strongest correlates of the two goal Bandura (1986) noted that a self-efficacy measure and the
measures and present performance. (As with Table 2 in Study behavioral test with which it is being correlated should be ad-
1, correlations that differ by more than . 15 are generally statis- ministered closely in time. In Study 1, self-efficacy was assessed
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

tically different. Specific tests are available from Cynthia Lee.) after students had taken their first examination but before they
The bottom half of Table 4 shows partial correlations be- had received their grade. In Study 2, self-efficacy was assessed
tween self-efficacy and the outcome measures, controlling for after the practice trials of a brainstorming task. However, be-
assigned goal. Note that the underlying pattern of correlations cause of the nature of the brainstorming task, immediate objec-
is remarkably similar, although differences among the corre- tive feedback on task performance was available in Study 2; this
lations are not as pronounced. was less so in Study 1, because subjects had not yet received
their examination scores. According to Kazdin (1978), in the
Discussion case of Study 2, completing self-efficacy items (especially when
the items are hierarchically presented) may sensitize subjects to
These two studies demonstrate that the five self-efficacy oper- the number of uses they should list on the brainstorming task.
ationalizations are highly correlated. Of the five operationaliza- Alternatively, task performance (listing uses of an object) may
tions of self-efficacy, the measures with respect to multiple per- alter subsequent self-efficacy perceptions because these two
formance levels (self-efficacy strength, self-efficacy magnitude, measures are administered at nearly the same time. Thus, the
and the two composite measures) showed higher convergent and net effect of this measurement context (self-efficacy items are
predictive validities than the one-item task-specific confidence framed in performance levels) and procedure may result in the
rating. These relationships appeared in both studies. On the ba- observed higher correlations between self-efficacy and perfor-
sis of these two empirical studies and the increased information mance in Study 2. Also note that, in Study 1, the self-efficacy
in the other four measures, we thus recommend that researchers subscales developed by Wood and Locke (1987) were not di-
refrain from using single-item measures of confidence as in- rectly framed in performance levels (i.e., levels of examination
dexes of self-efficacy. Also, the self-efficacy magnitude and scores) but in terms of task requirements essential for successful
strength measures appear to have generally weaker predictive class performance (e.g., class concentration, memorization, un-
validities and correlations than the self-efficacy composites derstanding, and note taking). Because the level of self-efficacy
have. Of the two composite measures, Self-Efficacy Composite was inferred from a different set of items than those used for
1 showed slightly stronger relationships with personal goals in behavioral or performance assessment, the correlation between
Study 2; but in most cases across both studies, the strength of self-efficacy and behavioral performance would decline. Thus,
the relationships of these composite measures with antecedents in addition to the timing of self-efficacy assessment, the mea-
and outcomes were similar. Because the operationalization of surement context and content of self-efficacy are also critical
these two composites was consistent with Bandura's (1986) con- and deserve closer attention from researchers.
ceptualization (incorporating both magnitude and strength in- In summary, it appears that there is convergent and predictive
formation) and because Composite 1 is less cumbersome to pp- validity across different types of self-efficacy measures and
erationalize, this may be the measure of choice. different performance contexts. The composite self-efficacy op-
It is interesting to speculate why the one-item confidence rat- erationalizations are consistent with Bandura's (1986) defini-
ing was more highly related to positive affectivity (and past per- tion and conceptualization, and the two present studies provide
formance) than were the other self-efficacy measures. The con- empirical evidence that these may be the measures of choice.
fidence rating asks about confidence "that you will obtain the Future studies should replicate our results in longitudinal field
score that you are trying for in your next exam." As one re- settings where performance feedback is perhaps less accurate,
viewer noted, this may well be interpreted as an item assessing or less timely, and where performance measures are often less
confidence in outcome expectancy rather than self-efficacy. Ac- quantifiable. Not all recent operationalizations of self-efficacy
cording to Bandura (1986), self-efficacy is a judgment of one's are equal. We suggest that researchers consider using and re-
capability to accomplish a certain level of performance, searching the relatively more information-rich (composite)
whereas an outcome expectation is a judgment of the likely con- measures in future empirical settings.
sequence such behavior will produce.
It is also suggestive that outcome expectancy does not occur References
independent of an individual's past performance. Notice also Bandura, A. (1986). Social foundations of thought and action: A social
that Watson et aJ. (J 988) assessed their positive affectivity mea- cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
SELF-EFFICACY MEASURES 369

Bandura, A., & Jourden, F. J. (1991). Self-regulatory mechanisms gov- perceived efficacy in social dilemmas. Journal of Experimental Social
erning the impact of social comparison on complex decision making. Psychology, 25, 287-313.
Journal of Personality and Social Psychology, 60, 941-951. Lee, C. (1988). Effects of goal setting and self-efficacy on female field
Bandura, A., & Wood, R. (1989). Effects of perceived controllability hockey team performance. International Journal of Sport Psychology,
and performance standards on self-regulation of complex decision 20, 147-161.
making. Journal of Personality and Social Psychology, 56, 805-814. Lee, C., & Bobko, P. (1992). Exploring the meaning and usefulness of
Barling, J., & Beattie, R. (1983). Self-efficacy beliefs and sales perfor- measures of subjective goal difficulty. Journal of Applied Social Psy-
mance. Journal of Organizational Behavior Management, 5,41-51. chology, 22, 1417-1428.
Berry, J. M., West, R. L., & Dennehey, D. M. (1989). Reliability and Locke, E. A., & Bryan, J. F. (1968). Grade goals as determinants of
validity of memory self-efficacy questionnaire. Developmental Psy- academic achievement. Journal of General Psychology, 79, 217-228.
chology, 25, 701-713. Locke, E. A., Frederick, E., Lee, C., & Bobko, P. (1984). Effect of self-
Bores-Rangel, E., Church, A. T., Szendre, D., & Reeves, C. (1990). Self- efficacy, goals, and task strategies on task performance. Journal of
efficacy in relation to occupational consideration and academic per- Applied Psychology, 69,241 -251.
formance in high school equivalency students. Journal of Counseling Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task
performance. Englewood Cliffs, NJ: Prentice Hall.
Psychology, 37, 407-418.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Matsui, T., Ikeda, H., & Ohnishi, R. (1989). Relations of sex-typed so-
Brown, S. D., Lent, R. W, & Larkin, K. C. (1989). Self-efficacy as a
This document is copyrighted by the American Psychological Association or one of its allied publishers.

cializations to career self-efficacy expectations of college students.


moderator of scholastic aptitude-academic performance relation-
Journal of Vocational Behavior, 35, 1-16.
ships. Journal of Vocational Behavior, 35, 64-75.
Matsui, T, & Tsukamoto, S. I. (1991). Relation between career self-
Cervone, D., & Palmer, B. W. (1990). Anchoring biases and the perse- efficacy measures based on occupational titles and Holland codes and
verance of self-efficacy beliefs. Cognitive Therapy and Research, 14, model environments: A methodological contribution. Journal of Vo-
401-416. cational Behavior, 38, 78-92.
Clement, S. (1987). The self-efficacy expectations and occupational McAuley, E., Wraith, S., & Duncan, T. E. (1991). Self-efficacy, percep-
preferences of females and males. Journal of Occupational Psychol- tions of success and intrinsic motivation for exercise. Journal of Ap-
ogy, 60, 257-265. plied Social Psychology, 21, 139-155.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation Peake, P. K., & Cervone, D. (1989). Sequence anchoring and self-effi-
analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. cacy: Primacy effects in the considerations of possibilities. Social
Frayne, C. A., & Latham, G. P. (1987). Application of social learning Cognition, 7, 31 -50.
theory to employee self-management of attendance. Journal of Ap- Podsakoff, P. M., & Farh, J. (1989). Effects of feedback sign and credi-
plied Psychology, 72, 387-392. bility on goal setting and task performance. Organizational Behavior
Gist, M. E., & Mitchell, T. R. (1992). Self-efficacy: A theoretical analysis and Human Decision Processes, 44, 45-67.
of its determinants and malleability. Academy of Management Re- Rebok, G. W, & Balcerak, L. J. (1989). Memory self-efficacy and per-
vim, 17, 183-211. formance differences in young and old adults: The effect of mne-
Gist, M. E., Schwoerer, C. E., & Rosen, B. (1989). Effects of alternative monic training. Developmental Psychology, 25, 714-721.
training methods on self-efficacy and performance in computer soft- Sherer, M., Maddux, J. E., Mercadante, B., Prentice-Dunn, S., Jacobs,
ware training. Journal of Applied Psychology, 74, 884-891. B., & Rogers, R. W. (1982). The self-efficacy scale: Construction and
Gould, D., Hodge, K., Peterson, K.., & Giannini, J. (1989). An explor- validation. Psychological Reports, 51, 663-671.
atory examination of strategies used by elite coaches to enhance self- Taylor, M. S., Locke, E. A., Lee, C., & Gist, M. E. (1984). Type A be-
efficacy in athletes. Journal of Sport and Exercise Psychology, 11, havior and faculty research productivity: What are the mechanisms?
128-140. Organizational Behavior and Human Performance, 34, 402-418.
Holahan, C. K., & Holahan, C. J. (1987). Life stress, hassles, and self- Watson, D., Clark, L., & Tellegen, A. (1988). Development and valida-
efficacy in aging: A replication and extension. Journal of Applied So- tion of brief measures of positive and negative affect: The PANAS
cial Psychology, 17, 574-592. scales. Journal of Personality and Social Psychology, 54, 1063-1070.
Jones, G. R. (1986). Socialization tactics, self-efficacy, and newcomers' Wood, R. E., & Bandura, A. (1989). Impact of conceptions of ability on
adjustments to organizations. Academy of Management Journal, 29, self-regulatory mechanisms and complex decision making. Journal
262-279. of Personality and Social Psychology, 56, 407-415.
Wood, R. E., & Locke, E. A. (1987). The relation of self-efficacy and
Kavanagh, D. J., & Bower, G. H. (1985). Mood and self-efficacy: Impact
grade goals to academic performance. Educational and Psychological
of joy and sadness on perceived capabilities. Cognitive Therapy and
Measurement, 47, 1013-1024.
Research, 9, 507-525.
Kazdin, A. E. (1978). Conceptual and assessment issues raised by self- Received March 9, 1992
efficacy theory. Advanced Behavioral Research Therapy, 1, 177-185. Revision received October 28, 1993
Kerr, N. L. (1989). Illusions of efficacy: The effects of group size on Accepted October 29, 1993

Vous aimerez peut-être aussi