Psychology
in
rhe Schools
Volume
2l.
October,
1984
THE EFFECT OF SUCCESS OR FAILURE ON SELF-EFFICACY
AND TASK PERSISTENCE OF CONDUCT-DISORDERED CHILDREN
ROBERT D. LYMAN, STEVEN PRENTICE-DUNN, DAVID
R.
WILSON, AND STEPHEN A. BONFILIO
The
University
ofAlabama
Twenty-four children with conduct disorders were familiarized with a pattern-
matching task, and their self-efficacy expectations for this task were assessed. Subjects
were then given either “Success”
or
“Failure” feedback for a series
of
similar
problems, and their self-efficacy was reassessed. Finally, task persistence was
evaluated by allowing subjects to attempt as many additional pattern-matching
problems as they wished, with all subjects receiving failure feedback. Results indicated
that, although the two groups of subjects were initially similar, the success group’s
self-efficacy was significantly higher following feedback. The success group also
attempted significantly more problems while later receiving failure feedback, and
there was a significant positive correlation between task persistence and the second
self-efficacy rating. These findings were related to Bandura’s self-efficacy theory and
to educational programming for conduct-disordered children.
Recently, psychology has witnessed a proliferation of theories and therapies in
which cognitive processes are assigned a central role (e.g., Mahoney
&
Arnhoff, 1978).
Among these is Bandura’s (1977, 1982) self-efficacy theory. The key proposition of self-
efficacy theory is that an individual’s belief in his/her ability to execute a behavior cor-
rectly determines what behaviors will be initiated, how much effort will be expended
in
their performance, and how long an individual will persist in a behavior when success is
not forthcoming.
Among the factors cited by Bandura (1977) as potentially affecting the strength of
these self-efficacy expectations are success
or
failure experiences. Brown and Inouye
(1 978) demonstrated that vicarious failure experiences decrease self-efficacy. College
students observing a model similar to themselves
in
age and sex who repeatedly failed on
an anagram task suffered decrements in their own perceived competence on the same
task. In addition, self-efficacy was found to be predictive of persistence on the insoluble
task; the greater the perceived self-efficacy, the longer the subject persisted. Bonfilio
(198 1) found that college students experiencing success reported significantly stronger
belief
in
their ability to perform a stress reduction procedure correctly than did students
experiencing failure. Bonfilio also reported a significant positive correlation between self-
efficacy and task persistence.
Bandura’s initial work on self-efficacy (Bandura, Adams,
&
Beyer, 1977) dealt
with
adults’ approach behavior toward phobic objects. Few studies have investigated self-
efficacy and achievement behavior
in
children. Schunk (198 1) found that children with
mathematics deficits exposed to adult models
or
written instructions designed to increase
personal mastery demonstrated increments in self-efficacy, achievement, and task per-
sistence relative to untreated controls. Subsequent analyses indicated a positive correla-
tion between self-efficacy and task persistence. Schunk (1982) also found that feedback
attributing past achievement to effort led to increased self-efficacy beliefs in these
children, while no feedback
or
feedback concerning future achievement had no effect.
Keyser and Barling (198
1)
found no relationship between academic accomplishments
and self-efficacy expectations
in
normal elementary school children, although they did
I
Reprint requests should be sent to Robert D. Lyman, Dept.
of
Psychology, Brewer-Porch Children’s
Center, University
of
Alabama, University, AL
35486.
516