You are on page 1of 13

Journal of Personality and Social Psychology Copyright 1981 by the American Psychological Association, Inc.

1981, Vol. 41, No. 3, 586-598 0022-3514/81 /4103-0586S00.75

Cultivating Competence, Self-Efficacy, and Intrinsic

Interest Through Proximal Self-Motivation
Albert Bandura and Dale H. Schunk
Stanford University
The present experiment tested the hypothesis that self-motivation through prox-
imal goal setting serves as an effective mechanism for cultivating competencies,
self-percepts of efficacy, and intrinsic interest. Children who exhibited gross
deficits and disinterest in mathematical tasks pursued a program of self-directed
learning under conditions involving either proximal subgoals, distal goals, or no
goals. Results of the multifaceted assessment provide support for the superiority
of proximal self-influence. Under proximal subgoals, children progressed rapidly
in self-directed learning, achieved substantial mastery of mathematical opera-
tions, and developed a sense of personal efficacy and intrinsic interest in arith-
metic activities that initially held little attraction for them. Distal goals had no
demonstrable effects. In addition to its other benefits, goal proximity fostered
veridical self-knowledge of capabilities as reflected in high congruence between
judgments of mathematical self-efficacy and subsequent mathematical perfor-
mance. Perceived self-efficacy was positively related to accuracy of mathematical
performance and to intrinsic interest in arithmetic activities.

Much human behavior is directed and sus- self-motivation relies on the intervening pro-
tained over long periods, even though the cesses of goal setting and self-evaluative re-
external inducements for it may be few and actions to one's own behavior. This form
far between. Under conditions in which ex- of self-motivation, which operates largely
ternal imperatives are minimal and discon- through internal comparison processes, re-
tinuous, people must partly serve as agents quires personal standards against which to
of their own motivation and action. In social evaluate ongoing performance. By making
learning theory (Bandura, 1977b, in press), self-satisfaction conditional on a certain
self-directedness operates through a self sys- level of performance, individuals create self-
tem that comprises cognitive structures and inducements to persist in their efforts until
subfunctions for perceiving, evaluating, mo- their performances match internal stan-
tivating, and regulating behavior. dards. Both the anticipated satisfactions for
An important, cognitively based source of matching attainments and the dissatisfac-
tions with insufficient ones provide incen-
This research was supported by Public Health Re- tives for self-directed actions.
search Grant M-5162 from the National Institute of Personal goals or standards do not auto-
Mental Health. We are deeply indebted to the many matically activate the evaluative processes
people who assisted us in this project: Ruthe Lundy and that affect the level and course of one's be-
Jack Gibbany of the Palo Alto Unified School District
arranged the necessary research facilities. School prin- havior. Certain properties of goals, such as
cipals Jerry Schmidt, Gene Tankersley, John Tuomy, their specificity and level, help to provide
Roger Wilder, and their staffs offered whatever help clear standards of adequacy (Latham &
was needed to facilitate the research. Jamey Friend and Yukl, 1975; Locke, 1968; Steers & Porter,
Barbara Searle of the Stanford Institute for Mathe-
matical Studies in the Social Sciences furnished us with 1974). Hence, explicit goals are more likely
invaluable information on mathematical subfunctions, than vague intentions to engage self-reactive
which served as the basis for the self-instructional ma- influences in any given activity. Goal prox-
terial. Finally, we owe a debt of appreciation to Debby imity, a third property, is especially critical
Dyar and Linda Curyea for their generous and able because the more closely referential stan-
assistance in the conduct of the experiment.
Requests for reprints should be sent to Albert Ban- dards are related to ongoing behavior, the
dura, Department of Psychology, Building 420, Jordan greater the likelihood that self-influences
Hall, Stanford University, Stanford, California 94305. will be activated during the process. Some

suggestive evidence exists that the impact of erned, at least in part, by goal proximity.
goals on behavior is indeed determined by Most of the activities that people enjoy doing
how far into the future they are projected for their own sake originally had little or no
(Bandura & Simon, 1977; Jeffery, 1977). interest for them. Young children are not
In the social learning view, adopting prox- innately interested in singing operatic arias,
imal subgoals for one's own behavior can playing tubas, deriving mathematical equa-
have at least three major psychological ef- tions, writing sonnets, or propelling heavy
fects. As already alluded to, such goals have shotput balls through the air. However,
motivational effects. One of the propositions through favorable continued involvement,
tested in the present experiment is that self- almost any activity can become imbued with
motivation can be best created and sustained consuming significance.
by attainable subgoals that lead to larger One can posit at least two ways in which
future ones. Proximal subgoals provide im- proximal goals might contribute to enhance-
mediate incentives and guides for perfor- ment of interest in activities. When people
mance, whereas distal goals are too far re- aim for, and master, desired levels of per-
moved in time to effectively mobilize effort formance, they experience a sense of satis-
or to direct what one does in the here and faction (Locke, Cartledge, & Knerr, 1970).
now. Focus on the distant future makes it The satisfactions derived from subgoal at-
easy to temporize and to slacken efforts in tainments can build intrinsic interest. When
the present. performances are gauged against lofty, dis-
Proximal subgoals can also serve as an tal goals, the large negative disparities be-
important vehicle in the development of self- tween standards and attainments are likely
percepts of efficacy. Competence in dealing to attenuate the level of self-satisfaction ex-
with one's environment is not a fixed act or perienced along the way.
simply knowing what to do. Rather, it in- Conceptual analyses of intrinsic interest
volves a generative capability in which com- within the framework of both self-efficacy
ponent skills must be selected and organized theory (Bandura, 1981) and intrinsic moti-
into integrated courses of action to manage vation theory (Deci, 1975; Lepper & Greene,
changing task demands. Operative compe- 1979) assign perceived competence a me-
tence thus requires flexible orchestration of diating role. A sense of personal efficacy in
multiple subskills. Self-efficacy is concerned mastering challenges is apt to generate
with judgments about how well one can or- greater interest in the activity than is self-
ganize and execute courses of action re- perceived inefficacy in producing competent
quired to deal with prospective situations performances. To the extent that proximal
containing many ambiguous, unpredictable, subgoals promote and authenticate a sense
and often stressful elements. Self-percepts of causal agency, they can heighten interest
of efficacy can affect people's choice of ac- through their effects on perception of per-
tivities, how much effort they expend, and sonal causation.
how long they will persist in the face of dif- Investigations of intrinsic interest have
ficulties (Bandura, 1977a; Brown & Inouye, been concerned almost exclusively with the
1978; Schunk, 1981). effects of extrinsic rewards on interest when
Without standards against which to mea- it is already highly developed. Although re-
sure their performances, people have little sults are somewhat variable, the usual find-
basis for judging how they are doing or for ing is that rewards given regardless of qual-
gauging their capabilities. Subgoal attain- ity of performance tend to reduce interest,
ments provide indicants of mastery for en- whereas rewards for performances signifying
hancing self-efficacy. By contrast, distal competence sustain high interest (Boggiano
goals are too far removed in time to provide & Ruble, 1979; Enzle & Ross, 1978; Lepper
sufficiently clear markers of progress along & Greene, 1979; Ross, 1976). The contro-
the way to ensure a growing sense of self- versy over the effects of extrinsic rewards on
efficacy. preexisting high interest has led to a neglect
The processes underlying the development of the issue of how intrinsic interest is de-
of intrinsic interest may similarly be gov- veloped when it is lacking. One of the present

study's aims was to test the notion that prox- Of the 25 pretest problems, 8 were similar in form
imal subgoals enlist the type of sustained and difficulty to some of the types of items used in the
subsequent treatment phase. To test for generalization
involvement in activities that build self-ef- effects, 17 problems required application of the various
ficacy and interest when they are absent. subtraction operations to problem forms that were more
Children who displayed gross deficits in complex than those children would encounter in the self-
mathematical skills and strong disinterest in instructional phase. For example, in treatment they
learned how to borrow once or twice from zero in, at
such activities engaged in self-directed most, four-column problems, whereas a generalization
learning over a series of sessions. They pur- item would require borrowing from three consecutive
sued the self-paced learning under condi- zeros in a six-column set. To add a further element of
tions involving either proximal subgoals, dis- complexity, all the test problems were cast in a form in
tal goals, or bids to work actively without which the minuend and the difference between the min-
uend and the subtrahend were provided so that children
any reference to goals. It was predicted that had to solve the subtrahend.
self-motivation through proximal subgoals Children were presented the set of 25 subtraction
would prove most effective in cultivating problems one at a time on separate pages and were in-
mathematical competencies, self-percepts of structed to turn each page over after they had solved
efficacy, and intrinsic interest in mathemat- the problem or had chosen not to work at it any longer.
The tester recorded the time spent on each problem.
ical activities. For reasons given earlier, dis- The measure of competence in subtraction was the num-
tal goals were not expected to exceed bids ber of problems in which the children applied the correct
to work actively in promoting changes. It subtraction operation.
was hypothesized further that strength of Pilot work in the development of the test procedures
self-efficacy would predict subsequent ac- revealed that children who do not fully comprehend sub-
traction operations fail to grasp the nature of their de-
curacy on mathematical tasks and level of ficiency because they faithfully apply an erroneous al-
intrinsic interest. gorithm. When presented with a subtraction problem,
they simply subtract the smaller number from the larger
one in each column regardless of whether the smaller
Method number is the minuend or the subtrahend. To address
this problem at the outset of the experiment proper,
Subjects after children completed the arithmetic pretest they
compared their solutions to the correct answers. How-
The subjects were 40 children of predominantly mid- ever, children performed the posttest without feedback
dle-class backgrounds, ranging in age from 7.3 to 10.1 of accuracy.
years, with a mean age of 8.4 years. There were 21 Since this study centered on motivational processes
males and 19 females distributed equally by age and by which competencies, perceived efficacy, and interest
sex across conditions. can be developed when they are lacking, children who
Children were drawn from six elementary schools. As solved more than four problems were excluded. The se-
an initial screening procedure, teachers identified chil- lected sample of children indeed exhibited gross deficits;
dren in their classes who displayed gross deficits in arith- one third could not solve a single problem, and another
metic skills and low interest in such activities. The pre- third could only solve one.
treatment procedures were administered to each of the The children's substantial quantitative deficiencies
identified children by one of two testers (a male and a were further confirmed by standardized measures of
female) to determine whether the children's arithmetic their mathematical ability obtained from the school dis-
skills were sufficiently deficient to qualify for the ex- trict on three subtests of the Metropolitan Achievement
periment. Test (Durost, Bixler, Wrightstone, Prescott, & Balow,
1970). They occupied the bottom percentile ranks in
computation (22), concepts (27), and problem solving
Pretreatment Measures (22). Children in the various treatment conditions did
not differ in this respect.
The study was presented to the children as a project Self-efficacy judgment. Before measuring perceived
aimed at gaining understanding of how arithmetic skills mathematical efficacy, children performed a practice
are acquired. To reduce further any evaluative concerns, task to familiarize them with the efficacy assessment
they were informed that the project was being conducted format. The tester stood at varying distances from the
in several schools and that their work would be treated children and asked them to judge whether they could
in full confidence. jump selected distances and to rate on a 100-point scale
Mathematical performance test. The performance the degree of certainty of their perceived capability. In
pretest consisted of 25 subtraction problems graded by this concrete way, children learned how to use numerical
level of difficulty. The test problems, which ranged from scale values to convey the strength of their self-judged
two to six columns, were specifically designed so as to efficacy.
tap each of seven subtraction operations that were in- In measuring strength of mathematical self-efficacy,
cluded in the treatment phase of the study. children were shown, for 2 sec each, 25 cards containing

pairs of subtraction problems of varying difficulty. This working on their own was underscored, they rarely
brief exposure was sufficient to portray the nature of sought the experimenter's attention during the sessions.
the tasks but much too short to even attempt any so- The self-directed study was conducted on consecutive
lutions. After each sample exposure, children judged school days. At the end of each session, children marked
their capability to solve the type of problem depicted where they had stopped and simply resumed work at
and rated privately the strength of their perceived ef- that point in the subsequent session.
ficacy on a 100-point scale, ranging in 10-unit intervals The situational arrangement for the instructional
from high uncertainty through intermediate values of phase of the study was designed to leave the initiative
certainty to complete certitude. The higher the scale to the children, thus allowing leeway for self-directed-
value, the stronger the perceived self-efficacy. The mea- ness and self-motivation to exert their effects. By work-
sure of strength of self-efficacy was obtained by dividing ing independently but in a group setting, none of the
the summed magnitude scores by the total number of children was the focus of attention. The seating ar-
problems. rangement and sequential entry precluded communi-
cation between the children. After delivering the in-
structions, the experimenter retired to a table away from
Self-Instructional Material the children and remained as unobtrusive as possible
throughout the sessions. By having children in different
Research conducted at the Stanford Institute of treatments pursue separately the self-directed learning
Mathematical Studies has shown that competence in in the same setting at the same time, social and situa-
subtraction requires several subskills (Friend & Burton, tional factors that might otherwise vary were compa-
1981). These include subtracting a number from a larger rable across treatment conditions.
one, subtracting zero, subtracting a number from itself,
borrowing once, borrowing caused by zero, borrowing
twice, borrowing from 1, and borrowing from zero.
Seven sets of instructional material were designed, Treatment Conditions
which incorporated the various subtraction operations.
The material was organized in such a way that children Children were assigned randomly to one of three
could work independently at their own pace over a series treatment conditions or to a nontreated control group.
of sessions. The instructions, format, and materials for the self-di-
The format of each instructional set was identical. rected study were identical across treatment conditions
The first page of each set contained a full explanation except for variations in goal setting.
of the relevant subtraction operation, along with two Proximal goals. For children in the proximal-goal
examples illustrating how the solution strategies are treatment, the experimenter suggested that they might
applied. The following six pages contained sets of prob- consider setting themselves a goal of completing at least
lems to be solved using the designated operation. Pre- six pages of instructional items each session. To give
testing showed that if children worked at a steady pace, some salience to a continuing goal orientation, the sug-
they could complete each self-instructional set in about gestion of proximal goals was made at the beginning of
25 min. the second session as well. There was no further mention
of goals thereafter.
Distal goals. For children assigned to the distal-goal
Procedure for Self-Directed Learning treatment, the experimenter suggested that they might
consider setting themselves the goal of completing the
One of three experimenters (one male and two fe- entire 42 pages of instructional items by the end of the
males) brought the children individually, at slightly seventh session, which comprised a total of 258 prob-
staggered times, into the study room, where they were lems.
seated in different locations, facing away from each In both treatment conditions, the goals were men-
other to preclude any visual contact. Both the experi- tioned suggestively rather than prescriptively so as to
menter and the schools from which the children were leave the goal-setting decision to the children. This mode
drawn were the same across treatment conditions. The of goal structuring was used for two reasons. First, it
entire set of instructional materials was placed face was designed to increase children's self-involvement in
down on the table. The children were informed that they the instructional task. Second, choice increases the level
could work on these subtraction problems for seven 30- of personal responsibility and commitment to goals.
min. sessions. No goals. A third group of children pursued the self-
In describing the procedure for self-directed learning, directed learning without any reference to goals. How-
the experimenter turned over the first page, which ex- ever, they were told to try to complete as many pages
plained the subtraction operation for the first six-page of instructional items as possible as they went along.
set. Children were told that whenever they came to a The reasons for including this particular condition were
page of instructions, they should bring it to the exper- twofold: to provide a control for the effects of self-di-
imenter who would read it to them. Then they should rected instruction alone and to equate the groups for the
solve, on their own, the subtraction problems contained social suggestion that they work productively.
on the succeeding pages. If children asked for further No treatment. A fourth group of children was ad-
assistance with the instructions, the experimenter simply ministered the full set of assessment procedures without
reread the relevant sections of the instructions but never any intervening exposure to the instructional material.
supplemented them in any way. Since the instructions This group provided a control for any possible effects
were self-explanatory and the importance of children of testing and concomitant classroom instruction.

Posttreatment Assessment no knowledge of the conditions to which the children

had been assigned.
The procedures used in the pretreatment phase of the After the experiment was concluded, all children, in-
study were readministered on the day following com- cluding the controls, pursued self-directed instruction
pletion of the fourth session. This intermediate point to completion under proximal subgoals to provide max-
was selected to gauge the effects of goal proximity on imal benefits for all participants.
the development of skill, self-efficacy, and intrinsic in-
terest within an identical length of time. Had children Results
been tested after completing the entire program of study,
the posttreatment changes would have been confounded
by variations in the amount of time different children No significant sex differences were found
required to complete the self-instruction. on any of the measures at either the pretest
Children's mathematical self-efficacy was measured or posttest assessments, since the sample was
at the end of treatment and after the posttest of sub- confined to children with gross arithmetic
traction performance. The self-efficacy scores obtained deficits. In the posttest assessment, children
at the end of treatment were used to gauge the value
of self-efficacy judgment in predicting subsequent arith- showed comparable gains in self-efficacy and
metic performance. Since posttest performance can af- arithmetic performance on generalization
fect perceived efficacy, the measure of self-efficacy ob- problems and on the types of items used in
tained following the arithmetic posttest was related to the treatment phase. Having mastered par-
the subsequent measure of intrinsic interest. ticular subtractive operations on simpler ex-
Each of the 25 pairs of efficacy assessment items,
which were the same as those used in the pretreatment emplars, children applied them accurately
assessment, corresponded in form and difficulty level to to more complex forms. The data were there-
a subtraction problem in the performance test but in- fore pooled across sex and class of item for
volved different sets of numbers. As noted previously, the primary analyses.
most of the test problems were more complex than the
ones included in the treatment phase of the study. Be- Analyses of variance were computed on
cause the test of self-efficacy tapped new applications the different sets of data, with phases of the
of cognitive operations, children had to rely on gener- experiment and treatment conditions rep-
alizable perceptions of their mathematical capabilities resenting the main factors. At the pretest
in making their efficacy judgments. phase, the groups did not differ on any of
A parallel form of the performance test used in the
pretest was devised for the posttreatment assessment of the measures. Significant intergroup differ-
mathematical competence. This eliminated any possible ences obtained in the posttreatment phase
effects due to familiarity with problems. Both forms were analyzed further, using the Newman-
were administered in a counterbalanced order to a sam- Keuls multiple-comparison method. Table
ple of 17 children who were not participants in the for- 1 shows the significance levels of the treat-
mal study. The alternate forms yielded highly compa-
rable scores (r = .87). ment effects, the changes achieved by chil-
Children's intrinsic interest in subtraction problems dren within each condition, and comparisons
was measured in a separate session scheduled the day between treatment conditions.
after the posttreatment assessment. The tester explained
that she/he had another task the children could do.
Their attention was then drawn to two stacks of 10 pages Perceived Self-Efficacy
each. One stack contained 60 subtraction problems of
varying levels of difficulty; the other stack contained The strength of children's perceived math-
rows of digit-symbol problems adapted from the Wechs- ematical efficacy at different phases of the
ler Intelligence Scale for Children (Wechsler, 1974). experiment is presented graphically in Fig-
The latter task involved filling in rows of empty squares
with symbols corresponding to the digits appearing ure 1.
above each square. Analysis of these data shows the main
The tester stressed that the children should feel free effect of treatment, F(3, 36) = 10.13, p <
to decide whether they wanted to work on one, or the .001, and the interaction between treatment
other, or both tasks. It was further emphasized that it
was up to them to decide how much time they wanted and experimental phases, F(6, 72) = 5.96,
to spend on each activity. The children worked alone p < .001, to be highly significant.
until 25 min. had elapsed. The number of subtraction Intragroup comparisons of changes in
problems the children solved under these permissive strength of self-efficacy, evaluated by the t
choice conditions constituted the measure of intrinsic test for correlated means, yielded no signif-
All of the assessment procedures were administered icant differences for children in the control
individually by the same tester in both phases of the group (Table 1). Those who had the benefit
study. To control for any possible bias, the testers had of proximal subgoals substantially increased

Table 1
Significance of Intergroup Differences and Intragroup Changes
Proximal Proximal Distal Distal
Proximal vs. no vs. vs. no vs. No goals
Measure vs. distal goals control goals control vs. control

Intergroup comparisons
(Newman-Keuls comparisons)
Strength of Self-
Post, <.05 <.05 <.01 ns <.05 IM
Post2 <.01 <.05 <.01 ns <.05 <.01
performance <.01 <.01 <.01 IM <.01 <.01
Easy problems IM IM IM ns ns ns
problems ns ns <.05 IM <.05 <.Q5
Intrinsic interest <.05 <,05 <.05 ns ns ns
Accuracy of
Appraisal of
Efficacy <.05 <.05 <.05 iw ns ns

Intragroup changes
0 tests)

Proximal Distal Control

Strength of Self-efficacy
Pre vs. Post, 4.69*** 2.93** 2.16* 0.01
Pre vs. Post2 5.90**** 2.15* 3.70*** 0.78
Posti vs. Post2 3.55*** 2.75** 1.18 1.12
Arithmetic Performance 12.62**** 3.17*** 4.27*** 1.01
Easy problems 0.72 0.26 0.32 4.12***
Difficult problems 4.57*** 1.41 1.76 3.34***

* p< .10. ** p < .05. *** p < .01. "p< .001.

their perceived self-efficacy and exhibited ioral posttest (Table 1). Children in the dis-
even further gains following the performance tal condition also exceeded the controls in
posttest. Children oriented toward distal self-efficacy, but they did not differ signifi-
goals displayed a moderate increase in self- cantly from those who set no goals for them-
efficacy but a small decline after the posttest. selves. Children in the latter condition judged
Self-directed learning without goals pro- their mathematical efficacy more highly
duced a modest increase at a borderline level than did the controls after but not before the
of significance. performance posttest.
In separate comparisons between treat-
ments, the proximal group exceeded all oth- Mathematical Performance
ers in strength of perceived self-efficacy, as Figure 1 presents the mean scores ob-
measured both before and after the behav- tained on the subtraction performance test


80 h o—o CONTROL

O 70
IL 60




Figure 1. The left panel shows the strength of children's self-percepts of arithmetic efficacy at the
beginning of the study (pretest), and before (Posti) and after (Postz) they took the subtraction posttest.
The right panel displays the children's level of achievement on the subtraction tests before and after the
self-directed learning.

by children in the various conditions. The solved only 5% of the problems in pretest
main effect of treatment was highly signif- and only 8% in posttest, remained grossly
icant, F(3, 36) = 12.80, p < .001, as was the deficient in this regard.
interaction between treatment and experi- Contrasts between the means of the dif-
mental phases, F(3, 36) = 12.55, p < .001. ferent treatment conditions show the chil-
Self-directed instruction promoted mas- dren in the proximal condition to be much
tery of subtractive operations in all three more skilled than those in the distal (p <
groups, whereas the controls remained at a .01), no-goals (p< .01), or control (p < .01),
loss on how to subtract numbers from each conditions. Children who pursued the self-
other (Table 1). In pairwise compari- learning with distal (/?<.01) or no goals
sons, children who had employed proximal (p < .01) were also more skilled than the
subgoals surpassed all the other groups in controls, but the former two groups did not
subtractive skills. Children who engaged in differ from each other.
self-directed learning either with distal or no
goals did not differ significantly from each Persistence
other, but both groups outperformed the
controls. Children who gain high self-efficacy
In the above measure, children received through skill acquisition solve problems
partial credit if they applied the appropriate readily and, therefore, need not spend much
subtractive operations but made a minor er- time on them. An aggregate measure of per-
ror in deriving or in recording the answer. sistence spanning the entire range of diffi-
The identical pattern of results is obtained culty is not too meaningful because long per-
on scores using a stringent criterion requir- severence times on hard problems are offset
ing perfect accuracy on all counts. In com- by rapid solutions of less difficult ones.
paring the children's subtractive skills before Changes in persistence were, therefore, an-
and after treatment, all three groups that alyzed separately for problems at two levels
engaged in self-directed instruction achieved of difficulty: The difficult set of problems
significant gains beyond the p < .001 level required two or more borrowing operations,
of significance; whereas the controls, who whereas the easier set involved either no bor-

rowing or, at most, only one smaller min- 14

Analysis of persistence on the easy arith-
metic items revealed no differences except
for the controls, who were significantly less
persevering (-31%) when tested again. £ 10
However, in the efforts expended on difficult t-
arithmetic items, the proximal children were
O 8
markedly more persistent after treatment
than before (+90%), the distal (+22%) and z
the no-goals children (+39%) were moder-
ately more persistent, whereas the controls
slackened their efforts (-27%) in the post-
test. This differential pattern of perseverence
yielded a highly significant Treatment X
Phases interaction, F(3, 36) = 5.67, p <
.005. Analyses of intragroup changes pre-
sented in Table 1 show that the increased
perseverence of the proximally self-moti- GOALS GOALS GOALS
vated children and the diminished effort of
Figure 2. Average number of subtraction problems chil-
the controls are highly significant. Multiple dren in the different conditions chose to solve when given
comparisons among groups in the posttest free choice of activities.
assessment disclose that children in each of
the treatment conditions, although not dif- quent productivity, since children in this con-
fering from each other, were all significantly dition were as prolific on the competing task
more persevering than were the controls and more so on the arithmetic one.
(Table 1).
Progress in Self-Directed Learning
Intrinsic Interest The average length of time it took children
The role of goal proximity in the devel- to complete each lesson was 21, 29, and 30
opment of intrinsic interest may be seen in min. for the proximal, distal, and no-goals
Figure 2. Analysis of variance of the number conditions, respectively. Thus, proximal self-
of subtraction problems that children chose motivators produced more rapid mastery of
to solve on their own yielded a significant the subject matter than did distal ones, F(l,
treatment effect, F(3, 36) = 3.57, p < .05. 27) = 3.94, p < .10, or self-instruction with-
Inspection of Table 1 shows that children in out goals, F(l, 27) = 5.44, p < .05.
the proximal subgoal condition exceeded all By the end of the four sessions, the per-
three comparison groups, which did not dif- centage of the total instructional material
fer from each other. Indeed, 90% of the chil- completed was 74% in the proximal condi-
dren who developed their arithmetic skill tion, 55% in the distal condition, and 53%
through the aid of proximal subgoals per- in the condition involving no goals. Although
formed subtraction problems under the free- proximal subgoals, compared to distal goals,
choice conditions; whereas only about 40% F(l, 27) = 3.66, p < .10, and no goals, F(l,
of the children in the other groups did so. 27) = 4.67, p < .05, fostered greater mastery
The involvement in arithmetic problems of subtractive operations, distal goals had no
displayed by the proximal children was not significant effect either on rate or level of
at the detriment of the competing activity. self-directed instruction.
Children in all groups performed a compa-
rable number of digit-symbol items and did Congruence Between Self-Efficacy
not differ in this respect (F = 0.77). It would Judgment and Performance
seem that experience with proximal self-mo- Congruence indices can be computed by
tivators enhances the total level of subse- comparing efficacy judgments at the end of

treatment with subsequent posttest perfor- tionship between theoretically relevant vari-
mance of subtraction problems of compa- ables. Product-moment correlations were
rable form and difficulty. In this procedure, calculated separately within groups, and
judgments of self-efficacy are dichotomized when they did not differ significantly, they
into positive and negative instances, based were averaged by means of an r to z trans-
on a selected cutoff strength value. Instances formation.
of congruence occur when children judged That skill acquisition builds self-efficacy
themselves capable of solving a given level receives support in the data. The more self-
of problem and, in fact, solved it or judged instructional material the children mastered,
themselves incapable and then failed the the stronger was their sense of mathematical
same class of problem. Mismatches between self-efficacy, r(28) = .42, /j<.01. Perfor-
efficacy judgments and performances (i.e., mances that are readily achieved suggest a
judged efficacy for failed items and judged higher level of self-ability than do analogous
inefficacy for solved items) represent in- attainments gained through slow, heavy la-
stances of incongruence. bor. Consistent with this expectation, the
Congruence indices were calculated sep- faster the children completed each lesson,
arately for efficacy cutoff values at different the more efficacious they judged themselves
levels of strength. When a low-strength value to be, r(28) = .32, /><.05. Standardized
is selected as the criterion of self-judged ef- measures of mathematical competence, based
ficacy, a weak sense of efficacy (e.g., 20) is on the Metropolitan Achievement Test, did
treated like complete certitude (100). Such not predict rate of skill acquisition or degree
a low criterion could produce artifactual of self-efficacy enhancement.
mismatches. If the criterion were set near Both instructional performance, r(28) =
the maximal strength value (80), reasonably .33, p < .05, and strength of self-efficacy,
high levels of self-judged efficacy (e.g., 70) K38) = .42,/><.01, were moderately related
would be defined as inefficacy. This too to the children's facility in using subtractive
would produce artifactual discrepancies. For operations. However, strength of self-effi-
the mathematical performances examined in cacy was a significant predictor of perfect
this study, an efficacy strength of 40, which arithmetic accuracy for the total sample,
reflects a moderate degree of assurance, pro- r(38) = .49, /x.OOl, and for the self-
vided the optimal cutoff criterion. instructed groups, r(28) = .40, p < .025,
Results of the congruence analysis dis- whereas past self-instructional performance
close that the conditions of treatment af- was only marginally related to faultless post-
fected the level of accuracy with which chil- test performance, r(28) = .25, p < .10.
dren appraised their mathematical efficacy, Skill acquisition speeds problem solving.
F(3, 36) = 3.06, p < .05. Children in the This factor attentuated the relationship be-
distal (54%), no-goals (51%), and control tween self-efficacy and persistence for the
(60%) conditions displayed moderate con- treated children who found most of the prob-
gruence between their self-judged efficacy lems readily soluble and, hence, had no need
and their performance. In contrast, children to spend much time on them. The influence
who developed their skills under proximal of perceived self-efficacy on perseverence is,
subgoals were highly accurate in their self- of course, best revealed on problems that
appraisals of efficacy (80%). Table 1 shows cannot be solved however hard one tries.
that in accurateness of self-appraisal, the Children who doubt their capabilities quit
proximally self-motivated children exceed sooner than those who believe they can even-
other experimental groups, which do not dif- tually master the task should they persevere.
fer significantly from each other. In most This condition obtains for the control chil-
instances the children erred by overestimat- dren whose marked arithmetic deficiencies
ing their capabilities. rendered most of the problems insoluble. For
this group, the stronger the children's self-
Correlational Analyses perceived efficacy, the longer they perse-
Correlational analyses were carried out to vered, K8) = .63, p = .025. For all children,
provide additional information on the rela- high perseverence was accompanied by high

performance attainments on the more dif- ical operations, and heightened their per-
ficult problems, r(38) = .51, p < .001. Even ceived self-efficacy and interest in activities
the easy problems were exceedingly difficult that initially held little attraction for them.
for the control children, and here too, per- Efforts to clarify how goal proximity op-
sistence was related to performance success, erates in self-regulatory mechanisms ordi-
K8) = .61, p < .001. narily present difficulties because even
The relationship of perceived self-efficacy though encouraged to set themselves distal
to intrinsic interest can be analyzed in sev- goals, people are prone to convert them into
eral ways. It may require at least moderately more aidful proximal ones. They simply
high self-efficacy to generate and sustain in- fractionate desired future accomplishments
terest in an activity, but interest is not much into attainable daily subgoals (Bandura &
affected by small variations above or below Simon, 1977). In the present experiment,
the threshold level. To test this threshold children could not transform distal into
notion, interest scores were correlated with proximal self-motivators because not know-
number of self-percepts of efficacy that ing how to divide, they could not partition
matched or exceeded an efficacy strength of the entire instructional enterprise into equiv-
40, which was previously shown to be the alent subunits. Results of the combined stud-
optimal criterion in the congruence analysis. ies attest to the motivating potential of prox-
The results disclose that the higher the level imal goals, whether they are suggested or
of self-efficacy at the end of the posttest, the spontaneously generated, or whether the
greater the interest shown in arithmetic ac- self-sustained behavior is tractable or very
tivities, r(38) = .27, p < .05. difficult to produce.
An alternative possibility is that intrinsic Judgment of mathematical self-efficacy
interest is linearly related to strength of self- by children just beginning to understand the
efficacy. Correlation of the latter measures requisite cognitive skills is no simple matter.
shows that variations in mean strength of With very brief exposure to sample items,
self-efficacy covaried with interest in the it is hard to discriminate among different
control and the no-goals conditions, K18) = levels of task difficulty. This is because the
.39, p < .05, but not in the goal-setting treat- complexity of some of the subtractive op-
ments. erations is not instantly apparent from what
Although interest was positively related is most readily observable. When complex
to self-percepts of efficacy derived from per- operations are imbedded in seemingly easy
formance attainments in treatment, it was problems, which is often the case, appear-
uncorrelated with the performance attain- ances can be quite misleading. In such sit-
ments themselves. However, the standard- uations, incongruities between perceived self-
ized measures of competence in mathemat- efficacy and action may stem from misper-
ical subfunctions emerged as significant ceptions of task demands as well as from
correlates. The more competent children faulty self-knowledge. Moreover, solving
were at mathematical computation, problems typically requires applying multi-
H38) = .42, p < .01, and problem solving, ple operations. Even if they were readily rec-
K38) = .58, p < .001, the more subtraction ognizable, judgment of personal capabilities
problems they completed in the free-choice for a given type of task is complicated if
situation. some of the constituent operations are thor-
oughly mastered and others are only par-
Discussion tially understood. Selective attention to mas-
tered elements highlights competencies;
Results of the present study confirm the whereas focus on what is less well understood
influential role of proximal self-motivators highlights shortcomings. Even equal atten-
in the cultivation of competence, self-per- tiveness to all aspects of the task will produce
cepts of efficacy, and intrinsic interest. Chil- some variance in judgments of self-efficacy,
dren who set themselves attainable subgoals depending on how much weight is given to
progressed rapidly in self-directed learning, the differentially mastered operations.
achieved substantial mastery of mathemat- Given these complexities, it is not sur-

prising that children sometimes overesti- primarily on how extrinsic incentives affect
mated their capabilities, especially on tasks high interest when it is already present
that appeared deceptively simple. However, rather than on how to develop it when it is
it is noteworthy that in addition to its other lacking. It is the latter problem that presents
benefits, goal proximity fosters veridical self- major challenges, especially when avoidance
knowledge of capabilities. Children who of activities essential for self-development
guided and judged their progress in terms reflects antipathy arising from repeated fail-
of proximal subgoals were highly accurate ure rather than mere disinterest. The present
in their self-appraisals. In contrast, skill de- findings lend support to the general thesis
velopment under conditions in which prog- that skills cultivated through proximal stan-
ress toward competence is somewhat ambig- dards of competency build interest in dis-
uous does not improve self-knowledge. valued activities. When progress is gauged
Despite the fact that children in the distal against distal goals, similar accomplish-
and no-goal conditions acquired new infor- ments may prove disappointing because of
mation about mathematics and applied it wide disparities between current perfor-
repeatedly, the accurateness of their self-ap- mance and lofty future standards. Conse-
praisals was no better than that of the con- quently, interest fails to develop, even though
trols, who did not have the benefit of skills are being acquired in the process.
substantial performance information for Perceived self-efficacy was accompanied
constructing their self-knowledge. by high-performance attainments and per-
The above results are consistent with pre- severence under conditions in which such a
vious findings that judgments of self-efficacy relationship would be expected to obtain.
are not simply reflectors of past performance Regardless of conditions of treatment, per-
(Bandura, 1977a, 1977b). Rather they re- sistency increased the likelihood of success.
flect an inferential process in which the self- There was some evidence to indicate that
ability inferences drawn from one's perfor- faultless arithmetic performance was better
mances vary, depending on how much weight predicted from self-efficacy than from be-
is placed on personal and situational factors havioral attainments in treatment. However,
that can affect how well one performs. The caution should be exercised in judging causal
evaluative standards against which ongoing contributions from correlations because some
performances are appraised constitute an of the measures reflect continuously inter-
additional factor that determines how well active processes rather than discrete sequen-
people judge their capabilities. tial ones. Consider, for example, the question
Results of microanalyses of congruence of directionality of influence in obtained re-
between self-efficacy judgment and perfor- lationships between treatment performance,
mance indicate that the optimal cutoff value posttreatment self-efficacy, and posttest per-
of efficacy strength varies across activities, formance. Self-efficacy judgments are un-
depending on the complexity and the variety confounded by future posttest performance,
of skills they- require. Performances that but it is highly unlikely that self-percepts of
draw on only a few skills reduce the likeli- efficacy played no role whatsoever in per-
hood of overestimating personal capabilities formance attainments during the self-di-
by overweighting a mastered subpart. Hence, rected learning. Judgments of one's capa-
lower efficacy strength values can be pre- bilities can affect rate of skill acquisition,
dictive of success (Bandura, 1977a). In ac- and performance mastery, in turn, can boost
tivities that depend on diverse subskills, self-efficacy in a mutually enhancing pro-
knowledge of some of them, especially the cess. It is not as though self-efficacy affects
more directly observable ones, raises the future performances in the posttest but does
level of assurance that one might be able to not affect earlier performances in the treat-
perform successfully. Consequently, some- ment phase.
what higher cutoff values of efficacy strength The causal contribution of perceived self-
become predictive of success. efficacy to performance is most clearly re-
Research on intrinsic interest has centered vealed in studies in which self-percepts of

efficacy are developed solely through vicar- poral lag effects warrant systematic inves-
ious or cognitive means that entail no overt tigation.
performance (Bandura & Adams, 1977;
Reference Note
Bandura, Adams, & Beyer, 1977; Bandura,
Adams, Hardy, & Howells, 1980; Bandura, 1. Bandura, A., Reese, L., & Adams, N. E. Micro-
analysis of action and fear arousal as a function of
Reese, & Adams, Note 1). Perceived self- differential levels of perceived self-efficacy. Unpub-
efficacy, instated symbolically, predicts well lished manuscript, Stanford University, 1981.
the pattern of performance successes and
failures on specific tasks. References
A further issue addressed in this research Bandura, A. Self-efficacy: Toward a unifying theory of
concerns the relationship between perceived behavioral change. Psychological Review, 1977, 84,
self-efficacy and intrinsic interest. It was 191-215. (a).
mainly children in the proximally self-mo- Bandura, A. Social learning theory. Englewood Cliffs,
N.J.: Prentice-Hall, 1977. (b)
tivated condition, all of whom felt highly Bandura, A. Self-referent thought: A developmental
efficacious, who displayed the notable level analysis of self-efficacy. In J. H. Flavell & L. Ross
of intrinsic interest. In contrast, children in (Eds.), Social cognitive development: Frontiers and
the other conditions generally expressed self- possible futures. Cambridge, England: Cambridge
doubts concerning their capabilities and University Press, 1981.
Bandura, A. The self and mechanisms of agency. In J.
showed little spontaneous interest in solving Suls (Ed.), Psychological perspectives on the self
arithmetic problems. Regardless of treat- (Vol. 1). Hillsdale, N.J.: Erlbaum, in press.
ment conditions, self-percepts of moderate Bandura, A., & Adams, N. E. Analysis of self-efficacy
to high strength were positively related to theory of behavioral change. Cognitive Therapy and
Research, 1977, /, 287-308.
interest. Bandura, A., Adams, N. E., & Beyer, J. Cognitive pro-
Intrinsic interest seems to covary most cesses mediating behavioral change. Journal of Per-
closely with the more long-standing indi- sonality and Social Psychology, 1977, 55, 129-139.
cants of actual or perceived competence, that Bandura, A., Adams, N. E., Hardy, A. B., & Howells,
is, the standardized measures of competence O. N. Tests of the generality of self-efficacy theory.
Cognitive Therapy and Research, 1980, 4, 39-66.
in mathematical subfunctions and perceived Bandura, A., & Simon, K. M. The role of proximal
self-efficacy in groups whose preexisting intentions in self-regulation of refractory behavior.
self-percepts were either unaltered or Cognitive Therapy and Research, 1977, 1, 177-193.
changed only marginally. These findings Boggiano, A. K., & Ruble, D. N. Competence and the
overjustification effect: A developmental study. Jour-
raise the interesting possibility that some nal of Personality and Social Psychology, 1979, 37,
temporal lag exists between newly acquired 1462-1468.
self-efficacy and corresponding growth of Brown, I., Jr., & Inouye, D. K. Learned helplessness
interest. The link between boosts in per- through modeling: The role of perceived similarity in
ceived self-efficacy and sustained involve- competence. Journal of Personality and Social Psy-
chology, 1978, 36, 900-908.
ment in challenging activities is now well Condiotte, M. M., & Lichtenstein, E. Self-efficacy and
established across a wide range of behavioral relapse in smoking cessation programs. Journal of
domains (Bandura, 1981; Brown & Inouye, Consulting and Clinical Psychology, (in press).
1978; Weinberg, Gould, & Jackson, 1979; Deci, E. L. Intrinsic motivation. New York: Plenum,
Schunk, 1981; Condiotte & Lichtenstein, in Durost, W. N., Bixler, H. H., Wrightstone, J. W., Pres-
press). But it may require mastery experi- cott, G. A., & Balow, I. H. Metropolitan Achievement
ences over a period of time before the self- Tests: Elementary form F. New York: Harcourt
efficacy derived from progressive successes Brace Jovanovich, 1970.
creates strong interest in activities that were Enzle, M. E., & Ross, J. M. Increasing and decreasing
intrinsic interest with contingent rewards: A test of
disvalued or even disliked. If, in fact, effects cognitive evaluation theory. Journal of Experimental
follow such a temporal course, then in- Social Psychology, 1978, 14, 588-597.
creased interest would emerge as a later Friend, J., & Burton, R. Teacher's guide: Diagnostic
rather than as an instant consequent of en- testing in arithmetic: Subtraction. Palo Alto: Xerox
Palo Alto Research Center, 1981.
hanced self-efficacy. Because of the theoret- Jeffery, K. M. The effects of goal-setting on self-mo-
ical import of the link between self-efficacy tivated persistence. Unpublished doctoral disserta-
and interest, both the threshold and the tem- tion, Stanford University, 1977.

Latham, G. P., & Yukl, G. A. A review of research on directions in attribution research. Hillsdale, N.J.:
the application of goal setting in organizations. Acad- Erlbaum, 1976.
emy of Management Journal, 1975, 18, 824-845. Schunk, D. Modeling and attributional effects on chil-
Lepper, M. R., & Greene, D. Overjustification research dren's achievement: A self-efficacy analysis. Journal
and beyond: Toward a means-ends analysis of intrin- of Educational Psychology, 1981, 73, 93-105.
sic and extrinsic motivation. In D. Greene & M. R. Steers, R. M., & Porter, L. W. The role of task-goal
Lepper (Eds.), The hidden costs of reward. Hillsdale, attributes in employee performance. Psychological
N.J.: Erlbaum, 1979. Bulletin, 1974, 81, 434-452.
Locke, E. A. Toward a theory of task motivation and Wechsler, D. Wechsler intelligence scale for children:
incentives. Organizational Behavior and Human Per- Form R. New York: The Psychological Corporation,
formance, 1968, 3, 157-189. 1974.
Locke, E. A., Cartledge, N., & Knerr, C. S. Studies of Weinberg, R., Gould, D., & Jackson, A. Expectations
the relationship between satisfaction, goal setting, and and performance: An empirical test of Bandura's self-
performance. Organizational Behavior and Human efficacy theory. Journal of Sport Psychology, 1979,
Performance, 1970, 5, 135-158. /, 320-331.
Ross, M. The self perception of intrinsic motivation. In
J. H. Harvey, W. J. Ickes, & R. F. Kidd (Eds.), New Received June 16, 1980 •