ersonality factors, indud.
ing self-esteem and self-
Self-Evaluation and Self-
Concept of Adolescents
With Physical
Disabilities
Gillian A. King, Izabela Z. Shultz,
Kathleen Steel, Michelle Gilpin,
Tarnzin Cathers
Key Words: adolescents. self-concept
Fifty-three adolescents aged 14 to 18 years with diag-
noses of cerebral palsy (n = 27), cleft lip or palate or
both (n = 17), or spina bifida (n = 9) took part in
this study examining their selfesteem, selfconcept,
selfacceptance, social selfefficacy, and values, as
measured by standardized instruments. Comparisons
were made separately for males and females with
nonns developed for adolescents without disabilities.
Significant differences were found only on several as-
pects of self-concept: females with physical disabilities
were lower in perceived social acceptance, athletic
competence, and romantic appeal than the nonnative
sample, and males with physical disabilities were low-
er in perceived scholastic competence, athletic compe-
tence, and romantic appeal. in addition, social self-
efficacy was found to be a significant predictor of
both independence and persistence in adolescents
with disabilities, who were sigmjicantly less inde-
pendent and persistent than were nonnative samples.
The discussion focuses on the usefulness of the findings
regarding social self-efficacy and the implications of
the findings for occupational therapists.
Gillian A. King, PhD, is Research Coordinator, Thames Valley
Children's Centre, 779 Base Line Road East, London, Ontario,
Canada, N6C 5Y6.
Izabela Z. Shultz, PhD, C Psych, is a Psychologist, Thames Valley
Children's Centre, London, Ontario, Canada.
Kathleen Steel, MSc, is a doctoral candidate, Department of
Epidemiology and Biostatistics, University of Western Ontario,
Ontario, Canada.
Michelle Gilpin, MA, is a Psychometrist, Thames Valley Chil-
dren's Centre, London, Ontario, Canada.
Tamzin Cathers, BA, is a Research Assistant, Thames Valley
Children's Centre, London, Ontario, Canada.
This article was accepted for publication August 12, 1992.
132
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
P concept, are among the main factors thought to be
associated with psychological problems in adoles-
cents (Anderson, Clarke, & Spain, 1982). Personality is
also considered a protective factor that can reduce the
likelihood of adjustment problems (Garmezy, 1983).
How adolescents with physical disabilities value them-
selves (their self-esteem) and view themselves (their self-
concept) may therefore be key predictors of their adjust-
ment and future life success (Kapp-Simon, 1986). In fact,
self-esteem, which is defined as a generalized feeling of
self-acceptance, goodness, and worthiness (Crocker &
Major, 1989), is considered a central aspect of psychologi-
cal functioning (Crocker & Major, 1989; Taylor & Brown,
1988) and has been shown to be related to a host of
variables including general satisfaction with one's life
(Diener, 1984).
Little information is available on the self-esteem and
self-concept of adolescents with physical disabilities (An-
derson et aI., 1982; Magill & Hurlbut, 1986). In the pres-
ent study, we examined both the self-evaluations (self-
esteem, social self-efficacy, and self-acceptance) and self-
concept (Le., perceptions of self on specific dimensions
such as academic ability and physical appearance) of ado-
lescents with cerebral palsy, spina bifida, and cleft lip or
palate or both in comparison to normative data. We also
tested several hypotheses about the relation between
these aspects of the self and independence (i .e., resisting
the influence of others) and persistence at tasks and
goals, which are considered important determinants of
academic and vocational success (Abramson, Ash, &
Nash, 1979).
Literature Review
There have been few empirical studies of the self-esteem
and self-concept of adolescents with cerebral palsy (Ma-
gill & Hurlbut, 1986), spina bifida (Campbell, Hayden, &
Davenport, 1977; Lord, Varzos, Behrman, Wicks, & Wicks,
1990; Murch & Cohen, 1989; Pearson, Carr, & Halliwell,
1985), or cleft lip or palate or both (Brantley & Clifford,
1979; Kapp, 1979; Richman, 1983; Starr, 1978). Several
studies have examined mixed subject groups consisting
of adolescents with different types of physical disabilities
or adolescents with chronic illnesses (Harper & Richman,
1978; Kellerman, Zeltzer, Ellenberg, Dash, & Rigler, 1980;
Pless, Cripps, Davies, & Wadsworth, 1989; Tavormina,
Kastner, Slater, & Watt, 1976). The majority of these stud-
ies have employed standardized measures of self-esteem
and self-concept but have used convenience samples
from clinic populations, which limits their generaliza-
bility.
These studies provide little evidence to support the
common assumption that adolescents with disabilities
have low self-esteem (see Bryan & Herjanic, 1980; Minde,
1978; Strax, 1988) or poor self-concepts (see Abramson et
February 1993, Volume 47, Number 2
 aI., 1979). These findings parallel conclusions drawn by
Wright (1960), who reviewed early studies on the adjust-
ment of persons with physical disabilities, and Crocker
and Major (1989), who reviewed research on the self-
esteem of stigmatized groups in general. Adolescents
with spina bifida did not appear to differ in self-esteem
from a comparison group of adolescents without disabil-
ities (Murch & Cohen, 1989), nor did they differ signifi-
cantly from matched controls on self-concept or overall
self-esteem (Campbell et aI., 1977). MagiJ! and Hurlbut
(1986) found no significant differences in overall self-
esteem between a group of adolescents with cerebral
palsy and a matched control group of adolescents without
disabiJities. In a study of adolescents with cleft lip or
palate or both, Starr (1978) found no significant differ-
ences in their self-esteem compared with that of a
matched control group. Brantley and Clifford (1979) re-
ported that adolescents with clefts had, in fact, greater
self-esteem and self-confidence than did comparison
groups of adolescents without disabilities and obese ado-
lescents. Kapp (1979) found no significant difference in
the global self-concept of adolescents with cleft lip or
palate and that of a group of matched controls.
In her review of the literature on the psychosocial
effect of chronic illness on children, La Greca (1990) con-
cluded that "the most striking and parsimonious conclu-
sion that can be drawn from this considerable data base is
that children with chronic or life-threatening illness ... do
not differ substantially from healthy youngsters in terms
of disease-specific personality patterns or prevalence of
severe emotional disorders" (p. 286). Although our focus
is on the self-evaluation and self-concept of adolescents
with physical disabilities, it appears that a similar general
conclusion is warranted. In fact, according to Crocker and
Major (1989), there is little empirical evidence that mem-
bers of any stigmatized group have low global self-
esteem. They proposed that a variety of protective
mechanisms is used to bolster self-esteem (e.g., persons
may attribute negative feedback to the fact that they be-
long to a stigmatized group rather than to faults that they
possess). Thus, both theoretically and empirically, there
is reason to question the common assumption that per-
sons with physical disabilities, as a group, are lower in
self-esteem than are persons without physical disabilities.
As Drotar (1981) has concluded with respect to child-
hood illness, it may be best to view a physical disability as
a stressor that, depending on its interactions with a host
of other variables, such as parental psychosocial charac-
teristiCS and family functioning, will have a varying effect
on the adolescent's self-esteem and self-concept.
Rather than attempting to determine whether ado-
lescents with disabilities, as a group, have lower self-
esteem and self-concept than do other adolescents, the
most productive next steps for researchers would be (a)
to determine the factors that have significant effects on
self-esteem and specific domains of self-concept within
The American Journal of Occupational Therapy
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
this population, and (b) to examine the effect of self-
esteem and self-concept on behavior, adjustment, and
key life outcomes (La Greca, 1990).
With regard to the former, it appears that gender
may be an important mediator of the effect of disability on
self-esteem and self-concept. In the group of studies re-
viewed above, those that employed matched control
groups (i.e., Campbell et aI., 1977; Kapp, 1979; Magill &
Hurlbut, 1986), although generally finding no overall dif-
ferences between disabled and control groups, did find
that boys and girls with disabilities differed from their
counterparts without disabilities on some subscales of
the standardized tests. Magill and Hurlbut (1986) found
girls with cerebral palsy to be significantly lower than all
other groups on their physical self-esteem. Kapp (1979)
found girls with clefts to be significantly lower than girls
without clefts on several subscales of the Piers-Harris
Children's Self Concept Scale (Piers & Harris, 1967), and
Campbell et a!. (1977) found significant differences be-
tween both boys and girls with spina bifida and matched
controls on several subscales of a self-image question-
naire. These findings, however, do not portray a consis-
tent pattern of differences for boys or girls, even on the
same measures. Both Campbell et a!. (1977) and Magill
and Hurlbut (1986) used the Tennessee Self-Concept
Scale (Fitts, 1965) and reported different findings. Differ-
ent samples may account for the inconsistencies or the
findings may be spurious. In fact, Campbell et al. (1977)
stated that, due to small sample sizes, their findings
"should be regarded as tentative and repeated on similar
groups" (p. 406), as well as on samples of adolescents
with other disabilities. The relatively small sample sizes
and large number of comparisons in the studies by Magill
and Hurlbut (1986) and Kapp (1979) make these
matched control studies vulnerable to spurious findings.
Study Rationale
The present study explored the role of gender in mediat-
ing the effect of disability on self-esteem and self-concept
by comparing means for males and females with disabil-
ities with normative data obtained from males and fe-
males without disabilities. A second aim was to extend
previous research, which has focused on self-esteem and
self-concept. Little attention has been directed toward
Other elements of the self that may be associated with
adolescents' successful adjustment to life. As Kapp (1979)
has stated, we need to take a more in-depth look at the
self-concept of persons with physical disabilities.
The aspects of the self examined were chosen on the
basis of a conceptual framework we developed to guide
our research. Five key components were examined: self-
esteem (Am I worthy?), self-concept (Who am P), self-
acceptance (Do I accept my disability?), social self-efficacy
(Am I competent in social situations?), and values (What
do I believe is worth doing?). The social self-efficacy and
133
 values of adolescents with physical disabilities are poten-
tially important variables that have not been examined in
previous studies. We believe that a sense of social compe-
tence and the values of goal orientation, achievement,
and decisiveness can greatly affect successful life out-
comes for adolescents with disabilities.
Social self-efficacy reflects a person's judgment
about whether he or she is capable of carrying out the
social tasks that underlie successful relations with others.
Self-efficacy is thought to be an important component of
social competence in both children (Rutter, 1985) and
adolescents (Connolly, 1989). Values are important as-
pects of the self (Rokeach, 1960) and may be instrumental
in determining personal and vocational activities (Gor-
don, 1967). Both immediate decisions and long-range
plans are influenced by value systems (Gordon, 1967).
Chronic disease and physical disabilities are thought to
affect values such as goal orientation (Kellerman et aI.,
1980; Werner & Smith, 1982). We examined the values of
goal orientation (the desire to have definite goals to work
toward), achievement (the desire to set the highest stan-
dards of accomplishment for oneself), and decisiveness
(the desire to think things through for oneself and make
decisions) .
As stated above, a second important direction for
research on the self-conceptions of adolescents with dis-
abilities is to examine the effect of self variables on be-
haviors predictive of successful adaptation to life (La
Greca, 1990). We examined the relation between aspects
of the self and the interpersonal style of adolescents with
disabilities - their characteristic ways of relating to other
people. More specifically, we were interested in examin-
ing the relative importance of self-esteem, social self-
efficacy, acceptance of disability, and values as determi-
nants of interpersonal independence and persistence. It
is commonly assumed that disabled persons lack inde-
pendence (Minde, 1978) and there are numerous refer-
ences to the issue of dependence for this group (Abram-
son et aI., 1979; Tavormina et aI., 1976), particularly
during adolescence (Bryan & Herjanic, 1980). Due to
parental overprotection, children with physical disabil-
ities, like those with chronic illness (Tavormina et a1.,
1976), may have particular difficulty in expressing their
own opinions and resisting the influence of others (Strax,
1988). Persistence at goals or tasks until they are complet-
ed is another area in which adolescents with disabilities
are assumed to be deficient because others often do
things for them and intervene when difficulties are expe-
rienced. Lack of independence and persistence likely will
have a negative effect on adolescents' academic and voca-
tional success (Abramson et aI., 1979).
Objectives
Using comparisons with normative data, this study exam-
ined whether male and female adolescents with physical
134
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
disabilities differed from male and female adolescents
without disabilities on selected self variables, most
of which have not been examined in the literature. We
also explored the relation between these aspects of the
self and independence and persistence in interpersonal
style. It was hypothesized that self-esteem, social self-
efficacy, acceptance of disability, and the values of goal-
orientation, achievement, and decisiveness would be
positively related to independence and persistence in in-
terpersonal style.
Method
Subjects
This study was a survey of the total population of clients
registered at Thames Valley Children's Centre (the major
rehabilitation center for children and adolescents with
physical disabilities in Southwestern Ontario) who met
the follOWing inclusion criteria: prior or current registra-
tion at the Centre; condition diagnosed as cerebral palsy,
spina bifida, or cleft lip or palate or both; age between 14
and 18 years; intelligence within normal range (IQ above
80 as determined by previous standardized psychometric
assessments); and residence within a 30-mile radius of
London, Ontario. Clients whose speech intelligibility was
less than 80% or who used an augmentative communica-
tion system, such as the Bliss Symbol System, were
excluded.
Ninety clients met the inclusion criteria. Of these, 78
were able to be contacted, whereas the others could not
be located. Of the 78 contacted, 53 (68%) consented to
participate. Various reasons were given by the 25 clients
who declined to participate: the travelling distance (n =
6); parent did not consent (n = 5); sensitive topic area
(n = 4); not interested (n = 8); busy schedule (n = 2).
Chi-square analyses revealed that the participants and
nonparticipants did not differ in terms of condition or
gender, X2 (2) = 2.9, n.s., and XZCl) = .7, n.s., respectively.
The 53 participants consisted of 16 males and 11
females with cerebral palsy, 3 males and 6 females with
spina bifida, and 12 males and 5 females with cleft lip or
palate or both (a total of 31 males and 22 females). The
mean age of the males was 16.4 years and the mean age of
the females was 15.7 years. The specific diagnoses for
each group were as follows: of the 27 adolescents with
cerebral palsy, 12 were hemiplegic, 9 were diplegic, and 6
were quadriplegic; of those with cleft lip or palate or both,
6 had cleft palate, 3 had cleft lip, and the remaining 8 had
both cleft lip and palate; all 9 participants with spina bifida
had hydrocephalus.
Procedure
All eligible subjects and their parents were contacted by
telephone by a research assistant who explained the pur-
pose of the study. A consent form and letter of explana-
February 1993, Volume 47, Number 2
 tion were then sent to interested clients. Subjects who
consented to participate attended either an individually
scheduled, 90-min testing session at Thames Valley Chil-
dren's Centre or completed the series of standardized
instruments in their home. Home visits were necessary
due to the limited mobility, transportation difficulties,
and physical needs of some of the subjects. Subjects re-
ceived $15 compensation for their participation.
Standardized Instruments
A battery of standardized instruments was used, all of
which are appropriate for adolescents. These instru-
ments have adequate psychometric properties and, for
the most part, provide norms based on samples of adoles-
cents from the general population. They were completed
by subjects in a standardized order. The instruments
were: (a) the Self-Perception Profile for Adolescents
(Harter, 1986), which proVides a measure of Global Self-
Worth and also measures perceived competence in eight
domains (Scholastic Competence, Social Acceptance,
Athletic Competence, Physical Appearance, Job Compe-
tence, Romantic Appeal, ConductIMorality, and Close
Friendship); (b) the Adolescent Social Self-Efficacy Scale
(Connolly, 1989), which elicits self-ratings of behavioral
effectiveness in problematic peer contexts; (c) the Atti-
tudes Towards Disabled Persons Scale (ATDP-Form B)
(yuker, Block, & ounng, 1966), which prOVides a mea-
sure of the degree to which the respondent perceives
disabled persons as being different from persons without
disabilities and has been widely used as a measure of
disabled persons' acceptance of their own disabilities
(Yuker et aI., 1966); (d) the Survey of Personal Values
(Gordon, 1%7), which provides measures of Goal Orien-
tation, Achievement, and Decisiveness, as well as Practical
Mindedness, Variety, and Orderliness; and (e) the Inter-
personal Style InventOlY (Lorr & Youniss, 1988), which
evaluates interpersonal style on 15 scales grouped under
five empirically-based factors - Interpersonal Involve-
ment, Autonomy (including the subscale Independent),
Stability, Socialization, and Self-Control (including the
subscale Persistent).
Ana~ysis Procedures
Comparisons with normative data. Sample scores
were compared, by gender, with norms for each scale
(after Anderson et aI., 1982; Tavormina et aI., 1976). The
normative samples consisted of students in grades 8 to 11
(for the Self-Perception Profiles for Adolescents) and high
school students (for the Adolescent Social Self-Efficacy
Scale, the Survey of Personal Values, and the Interperson-
al Style Inventory); scores by age or grade were not avail-
able for the ATDP. One-tailed, independent t-tests, with
pooled estimates of variance (as appropriate), were per-
formed to compare means of adolescents with disabilities
with the means reponed for the normative samples. Sep-
The American journal of Occupational Therapy
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
arate comparisons were made for males and females on
total scores or subscale scores, as appropriate.
Regression analyses. Two "all possible subsets" re-
gression analyses were performed, with Independence
and Persistence as the outcome variables. In each case,
Global Self-Worth (from the Self-Perception Profile for
Adolescents), Social Self-Efficacy, Acceptance of Disabil-
ity, and Goal-Orientation, Achievement, and Decisiveness
(from the Survey of Personal Values) were entered as
predictor variables.
Variables were entered into the analysis as T scores
(Independence, Persistence), raw scores (Goal Orienta-
tion, Achievement, Decisiveness, and Global Self-Worth),
or total scale scores (Social Self-Efficacy, Acceptance of
Disability). Gender and condition were controlled for in
the analyses through the use of dummy variables.
The predictor variables were plotted against the out-
come variables to look for nonlinear relations and to de-
termine whether transformation of predictor variables
was desirable. After selection of a model, residuals were
examined to detect outliers. Outliers were removed, indi-
vidually and in combination, to examine the stability of
the coefficients. All cases remained in the final analysis.
Final selection of a model was determined by a significant
increase in R 2 , in the presence of controlling variables.
Results
Differences Among DiagnostiC Croups
Before conducting the analyses of interest we examined
whether the three diagnostic groups (cerebral paJsy,
spina bifida, and cleft lip or palate or both) differed in
terms of their self-evaluations or self-concepts. On the
basis of the literature (Bryan & Herjanic, 1980; Wright,
1960), we did not expect to find that particular personal-
ity traits were characteristiC of one diagnostiC group more
than of anOther group.
A series of one-way analyses of variance were per-
formed on the subscales of the Self-Perception Profile for
Adolescents, the Survey of Personal Values, and the Inter-
personal Style Inventory, and on total Social Self-Efficacy
and Acceptance of Disability scores. Due to the large
number of statistical tests performed (n = 19) and our
desire to avoid spurious effects, alpha was set atp < .001
to control for type 1 error rate in this set of analyses (Kirk,
1968). No significant differences were found between the
diagnostic groups on the measures employed in this
study, except that adolescents with cleft lip or palate or
both had higher scores on Athletic Competence (M =
29) than did adolescents with cerebral palsy (M = 1.9) or
spina bifida (NI = 1.7, Fl2,50j = 9.2,p < .0005 [p <.05
for comparisons]).
Because there was only one significant difference
among diagnostic groups, the normative comparisons re-
poned here are based on all groups combined. It should
be noted that normative comparisons also were conduct-
135
 ed excluding the cleft lip or palate group. This was clone !\uitudes Towards Disabled Persons Scale
to investigate the possibility that adolescents with clefts Sdf-i\ccejlt:lncc ,\.\ 1063 lb9 29 110.2 2[5 54')
might raise the mean scores for the disabled group, thus F 1206 228 22 1135 220 ')49
obscuring differences between adolescents with and Sun--c)' of' Pusonal Values
without physical disabilities. The findings did not differ Coal Orien-
from those involving all diagnostic groups combined. t:ltiOll ;'vl 178 '5.1 26 181 5.5 1644
F 16.8 ').8 16 lH.2 5.0 1392
Achie\'elllellt M 1')1 3.6 26 165 46 1644
Comparisons with Normative Data r 17.3 '5. I 16 lSI '52 1392
Dl'cisivcncss M 14.7 51 26 145 52 1644
Means and standard deviations for adolescents with phys- F 14.'5 4.7 16 14.1 5.2 1392
ical disabilities were compared with normative data pro- Pracrical
vided by authors of the instruments, either in manuals or Minuedncss M 145 50 26 14.5 50 1644
r 139 4.H 16 14.0 ')1 1392
in articles addressing issues of construct validation (see Varier\' M 127 8.6 26 11.b 8.0 1644
Table 1). The actual size ofthe study sample differs across r II.') 9.1 16 134 78 1392
the various outcome measures due to missing data. We Oruerliness M 147 6.1 26 144 57 1644
F 14') 6.1 16 151 5.8 1392
also selected subjects in grades 9 to 13 on the measure of
Interpcrsonal Style Inventory
self-efficacy to obtain a more precis~ match with the nor-
mative sample. Alpha was set atp < .001, one-tailed to lndcpendcnt" M 44.6 144 28 500 100 225 -7.3''-
control for type 1 error rate. Several significant differ- F 4'53 12.b 21 500 100 425 -6.6':''''
Pcrsislcnl: 1 M 45.1 119 28 500 100 225 -7.7''*
ences were found. As can be seen in Table 1, both males F 411 137 21 ')00 100 423 -12S'"
and females with physical disabilities were significantly
No/e. M = male, r = female.
Higher scores renect higher puceivcd competence (Self-Perception
Profile Il,r Adolescents), higher evaluations or social self-efficacy (Ado-
Table 1 lescent Social Selr-Efficac\' Scale), gre,lter acceptance of disability (Atti-
Measures of Self-Evaluation, Self-Concept and tudcs Towards Disabled Persons Scale), grcater imporr'lIlce of specific
Interpersonal Style for Adolescents with Disabilities valucs (Survey of Personal Values), and grcatel' illliependcnce or pcrsis-
and for Normative Samples tence (lnrelTlcrsonal St)'le Inventory).
"Mean I-SCOI'CS arc presentcd 1'01' both study and normativc samplcs.
NormativL "p < .DOl. one-tailed.
Studv Sample Sampil's "':p < .cJOO'), one-railed
(adolescenr,; (adole,;ceJ1l';
with without
dis~bilities) disabilities) lower in Athletic Competence and Romantic Appeal than
jn~trLllncnt Gen- were the normative samples. In addition, disabled males
(:lIld ,;uhscail's) cler M SD n M SD 11 I-value,;
were significantly lower in Scholastic Competence and
Scif-PcI'ception Profile for Adolescents
disabled females were significantly lower in Social Accep-
Glohal Self-
Wonh M 51 '5 31 32 .6 461 tance_ Both males and females were significantly lower in
F 5D .6 22 30 .7 496 Independence and Persistence than were adolescents
Scholastic
without disabilities.
C'lllllJerenCe M 2.7 .7 31 30 .7 461 -3.2"
F 2.b .b 22 29 .7 496
Soci:ll Regression Analyses
Accepunce ;'\'I 5.0 .6 31 32 .6 461
r 2.7 .H 22 32 6 496 -3.5' Independence. Social self-efficacy was the only sig-
AthletiC nificant predictor of independence in interpersonal style.
C, ,mpctcnce M 2') 9 31 30 .H 461 -jH
r I.b 1.0 22 2'5 8 496 -39'
Its coefficient remained stable when various outliers were
PlwsicaJ removed. The variables indicating condition were not sig-
Appeal-ance M 7 '!
. 6 'II 28 .7 461 nificant, but were stable and were retained because of the
F 21 .7 22 .
? )- .H 496
.loll Com- unbalanced distribution of condition in the sample, in
pctencc M 30 .4 31 32 ') 461 spite of causing a reduction in the significance of the
F 3.1 ') 22 31 ') 496 equation (see Table 2).
Romantic
Appeal M 23 7 31 27 .6 461 - 32 / Gender was removed as a controlling variable be-
r 2.1 .7 22 26 7 496 - 3.9" cause both the sign and the magnitude of the coefficient
Conduu/ changed when outliers were removed, and therefore its
Moralit\' M 29 4 31 2.H 6 46
F 2.b ') 22 29 6 496 effect, in addition to being nonsignificant, was not inter-
C10sc pretable. Further analysis revealed that self-efftcacy and
Friendship ~'l 3.3 .6 31 31 8 461 independence were most strongly correlated for the
r 31 .H 22 34 .7 496
group with cerebral palsy, both males and females, and
Adolescent Social Self-Effkacy Scale
that they were uncorrelated for females with spina bifida
Sdf-Efficacv M 1188 IHO 28 1196 16.7 40 or clefts. The latter had very small numbers (six and five,
r 1191 232 18 12').1 197 47 respectively) in this sample.

136 February 1993, Volume 47, Number 2

Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms

 Persistence. As can be seen in Table 2, social self-
efficacy was again the only significant predictor of persis-
tence. Gender and condition were not significant but
were retained as controlling variables. The correlation
between subjects' scores on Independence and Persis-
tence was .32 (n = 49; P < .01).
Discussion
This study examined whether the self-evaluations and
self-concepts of adolescents with physical disabilities dif-
fered from normative data, and how these self variables
related to two key aspects of success in life, indepen-
dence and persistence at tasks, No differences were
found in the self-evaluations of male and female adoles-
cents with cerebral palsy, spina bifida, and cleft lip or
palate or both in comparison with normative data ob-
tained on persons without disabilities, However, adoles-
cents with disabilities differed from normative samples in
terms of several specific aspects of self-concept. Not sur-
prisingly, both males and females had lower perceived
athletic competence, Of greater interest are differences in
perceived romantic appeal (for both males and females),
scholastic competence (for males), and social acceptance
(for females), We also found that adolescents' evaluations
of their competence in social situations - their social self-
efficacy - was significantly related to both their inde-
pendence in interpersonal relations and their persistence
at tasks,
There were no differences between the sample of
adolescents with disabilities and normative samples on
measures of self-evaluation, We found no differences in
self-esteem, social self-efficacy, acceptance of disability,
and values for either males or females, Our findings
therefore extend previous research, which has generally
found no differences on global measures of self-esteem
(e,g., Campbell et aI., 1977; Kapp, 1979; Magill & Hurlbut,
1986; Murch & Cohen, 1989; Starr, 1978). We found no
evidence that adolescents with disabilities evaluate them-
Table 2
Regression Analyses Predicting Independence and
Persistence
Significance
Outcome Variahle Predictor Variahle CoeffiCient
Independencea Social self-efficacy 23
Cleft lip or palate
or both -2,08
Spina bifida -602
Intercept 1975
Persisrence b Social self-efficacy .26
Cleft lip or palate
or both 1.19
Spina bifida -3.88
-263
Gender
Intercept 1387
aOveral1 F = 3.38, P = .03, R2 = 18.4
bOveral1 F = 3,17, P = .02, R2 = 22.3
The American ]oumal of Occupational Therapy
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
selves differently in a global sense, whether it be in terms
of their self-esteem, self-acceptance, or perceived self-
efficacy in social situations. With respect to clinical impli-
cations, these findings indicate that clinicians should not
assume that adolescentS with physical disabilities will
have problems in self-esteem.
The generalizability of these findings concerning
self-evaluation needs to be considered, There were no
significant differences between the assessed sample and
the population from which they were recruited in terms
of diagnostic category or gender, and our participation
rate of almost 70% is quite high, However, because self-
administered measures were used, we did not include in
our study adolescents with cognitive impairments, who
might have lower self-esteem, It is also possible that those
adolescents with feelings of inadequacy were more likely
to decline to take part in the study. Our findings do
correspond, however, to those reported in the literature,
although these past studies have been subject to small
sample size and have not addressed the issue of statistical
power.
Statistical power refers to the ahiliry of the analysis to
detect true differences in the population under study,
Our study suffered from two factors that reduce power:
small sample size (differences must be larger to achieve
statistical significance when the samples being compared
are small) and the reduction in the criticalp-value (alpha)
necessitated by the large number of tests (which means
that larger differences are required before being declared
significant), Our power calculations indicate that we had
low power to detect differences that may truly exist.
If the difference of interest in this study is specified
as .5 standard deviations on each of the instruments (a
difference that Cohen [1988] defined as moderate or
"medium"), the power of our analysis can be calculated
The groups of males and females compared in Table 1 are
unequal, and therefore the harmonic mean must be used
in power calculations. For example, our analysis of the
Self-Perception Profile for Adolescents has the same row-
er as one that compares equal groups of 58 males and 42
females, which is 85% for males and 74% for females at
alpha (une.,i,lccl) = .05 (Cohen, 1988),62% and 47%, respec-
tively, at alpha (nne.sided) = ,01, but 35% and 22% ifalpha (one'
sided) is set to ,001 (Cohen, 1988). Similarly, for the analy-
oft ses of variance, the power to detect a moderate
01 (medium) difference among diagnostic groups if = .25,
n = 17, k = 3) for any instrument is 3% if alpha = .05,
.58 13% if alpha =01 (Cohen, 1988),
20
.06 Our aim was to be conservative and thiS strategy of
01
minimizing spurious findings led us to set stringent alpha
levels. This increases our confidence that the differences
76 reported here are true differences but decreases the
43
chance of finding true population differences of clinical
45
.21 significance. We are therefore unable to conclude firmly
that adolescents with disabilities and their peers without
disabilities do not differ in self-esteem.
137
 The sample sizes of thiS srudy are larger than those
of most published studies in the field (see e.g, Campbell
et aI., 1977; Magill & Hurlbut, 1986); all eligible subjects of
one major treatment center were approached for inclu-
sion in this srudy. To conclusively address whether, or to
what extent, adolescentS with and without disabilities dif-
fer in self-esteem requires a very large, multicentered
study with adequate sample size. The strengths of our
study in this regard have been (a) our explicit consider-
ation of the issue of statistical power, which has not been
addressed by previous research in this area, and (b) our
examination of other measures of self-evaluation (e.g.,
social self-efficacy and acceptance of disability) in addi-
tion to self-esteem.
We did find several significant differences in self-
concept that are noteworthy in light of our conservative
analyses. Both males and females with disabilities view
themselves as lacking in romantic appeal. This view may
be realistic, but nonetheless is an area of particular con-
cern and worry for this group. Anecdotal and clinical evi-
dence of this concern with dating and attractiveness to
the opposite gender is present in the literature (Bryan &
Herjanic, 1980), but the present study is the first to dem-
onstrate it empirically. Campbell et al. (1977) found male
adolescents with spina bifida to have poorer body images
than did matched controls, and Magill and Hurlbut (1986)
found girls with cerebral palsy to score lower than
matched controls on physical self-esteem. These con-
cerns with physical image may be related to our finding of
lowered perceived romantic appeal.
We also found that males with disabilities viewed
themselves as lower in scholastic competence than did
males without disabilities, a finding that has not been
previously reported in the literature. As well, females with
disabilities viewed themselves as lower in social accep-
tance than did female adolescents without disabilities,
which parallels Campbell et al.'s (1977) finding that ado-
lescent girls with spina bifida expressed more social con-
cerns, including isolation and loneliness, than did control
girls. Because the presence of good friendships has been
found to be a particularly important protective factor
against adjustment problems for adolescent girls (Rae-
Grant, Thomas, Offord, & Boyle, 1989), the lower per-
ceived social acceptance of females is of particular
concern.
Exploration of the relations between aspects of self
and independence and persistence reflects a general re-
search strategy that has clinical utility: the exploration of
the intrapersonal, interpersonal, familial, and environ-
mental factors associated with key life outcomes for ado-
lescents with disabilities. As expected, we found adoles-
cents with disabilities to be significantly lower in both
independence (the disposition to express one's opinions
and to resist the pressure of others to conform) and
persistence (the disposition to persist at tasks or goals
until they are completed, despite obstacles or difficul-
138
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
ties). We found that social self-efficacy was positively re-
lated to both increased independence and perSistence,
and that independence and persistence were only moder-
ately (yet significantly) correlated. These findings indicate
that social self-efficacy is an important predictor of two
relatively independent behavioral tendencies with impor-
tant implications for life success in adolescents with
disabilities
The concept of self-efficacy is receiving increased
attention in the literature on stress and coping (Moos &
Billings, 1982), although self-efficacy is usually conceptu-
alized as a personal resource used in handling adverse
environmental events, rather than in dealing with social
situations and difficulties. Persons with higher levels of
self-efficacy are thought to be more persistent in their
efforts to handle threatening situations, whereas those
low in self-efficacy tend to avoid such situations (Moos &
Billings, 1982). Our investigation of social self-efficacy in
adolescents with physical disabilities has indicated that
persons high in social self-efficacy are also more persis-
tent at tasks and goals. Concrete evidence for this relation
has not previously been demonstrated.
Although we found no overall differences between
adolescents with and without disabilities on social self-
efficacy, there are assuredly adolescents with social skill
deficits related to their feelings of inadequacy in dealing
with social situations (Shultz, Wan-Leeper, King, McGol-
drick, & Stanczyk, 1990). Certain adolescents, due to their
physical limitations and social isolation, will not have had
the mastery experiences in social situations that form the
basis for one's sense of efficacy.
The present research has indicated that exploration
of the social self-efficacy of adolescents with disabilities is
an important endeavor, because social self-efficacy is re-
lated to important behavioral tendencies for this group.
One useful research direction would be to examine
whether social self-efficacy can be enhanced by social
skills training programs. Because social self-efficacy is
based on self-expectations for interpersonal skill, these
programs could increase feelings of self-efficacy by pro-
Viding opportunities for successful experiences with
peers, reinforcing appropriate interpersonal behavior,
and encouraging realistic expectations. These strategies
are employed by many social skills programs, although
there are few programs for persons with disabilities
(Shultz et aI., 1990); furthermore, the effect of such pro-
grams on the skills and social competency of adolescents
with disabilities remains to be addressed. Social self-
efficacy might be an important outcome to include in
such effectiveness evaluations.
Implications for Occupational Therapy Practice
The findings have a number of implications for assess-
ment and intervention with children and adolescents with
physical disabilities. Occupational therapists can and do
February 1993, Volume 47, Number 2
 employ a wide variety of methods to facilitate persistence
and independence and decrease feelings of scholastic
incompetence, social nonacceptance, and lowered ro-
mantic appeal. Some of these methods are general tech-
niques that apply regardless of content, whereas others
are specific to a particular area of perceived difficulty.
Therapists should not assume that clients will have
self-esteem or self-concept difficulties and therefore
should assess for such concerns (Magill-Evans & Restall,
1991). Our findings challenge the widely held perception,
shared by therapists, that most persons with physical
disabilities have lower feelings of self-worth (Arnold &
Chapman, 1992). Our findings reinforce the need to use
a client-centered approach (Canadian Association of
Occupational Therapists [CAOT], 1991a) to determine
whether clients themselves, rather than parents or
teachers, have concerns related to self-evaluation or
self-perception. The newly developed Canadian Occupa-
tional Performance Measure (CAOT, 1991b) may be a
useful tool for assessment in this area as it enables the
client to identify areas of perceived needs and goals for
intervention. Our findings indicate that therapists assess-
ing clients with physical disabilities should pay particular
attention to concerns about romantic appeal, feelings of
scholastic incompetence (for boys) and feelings of social
nonacceptance (for girls).
General intervention strategies that are thought to
have positive effects on self-esteem and self-concept in-
clude involving the client in goal setting, encouraging
problem solVing and role playing around problem situa-
tions, and reinforcing behaviors related to the acqUisition
of competence. Teenagers may also need to be motivated
to acquire behavioral or social competenCies. On the ba-
sis of Harter's (1978) theory of competence motivation, it
is important to ensure that teenagers have an optimal
degree of challenge, are able to define the boundaries of
competence by experiencing failure, and are reinforced
for attempting to succeed. Perhaps most important, occu-
pational therapists need to empower clients to have more
control in therapy, to promote feelings of self-worth and
competence.
Specific suggestions can be made for interventions
designed to address particular types of difficulties. Per-
ceptions of lack of romantic appeal can be addressed
through problem solving around practical issues (such as
clothes and grooming); role playing (e.g., how to ask
someone for a date); discussing societal attitudes and
beliefs about physical attractiveness and sexuality (par-
ticularly with regard to those with physical disabilities);
and, as suggested by Magill and Hurlbut (1986), helping
teenagers redefine attractiveness. Depending on the rea-
sons for perceptions of lack of scholastic competence,
occupational therapists can target visual-perceptual, mo-
tor, or communication difficulties (Kielhofner, 1985),
teach organizational skills, or motivate teenagers to want
to do well by discussing the consequences of lack of
The Amen'can Journal of Occupational Therapy
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
academic interest. To foster social acceptance, therapists
should promote opportunities for girls to develop close
friendships. This may involve encouraging parents to al-
low their daughters to go out with others more frequently
or spend more time on the phone. Therapists might also
assist in setting up informal support groups for teenagers.
Adolescents with physical disabilities are described
as being more dependent on their parents than are ado-
lescents without disabilities, not only in a physical sense
(Blum, Resnick, Nelson, & St. Germaine, 1991), but be-
haviorally and socially as well (Bryan & Herjanic, 1980).
Therapists need to promote independence and auton-
omy early on and not wait for dependency issues to arise
in adolescence before addressing these issues with par-
ents and children. Useful techniques include problem
solving around issues of independence (e.g., difficulties
with parents and transportation problems), involving the
teenager in decision making and goal setting around
treatment, and training in assertiveness. A lack of persis-
tence may be tied in with dependency on others, because
children may be less likely to persist at tasks when they
know that parents will provide assistance when request-
ed. Therapists can promote persistence at tasks by dis-
cussing the issue with parents and helping them see the
longer-term, negative consequences of assisting their
child too readily.
Future research should attempt to replicate the find-
ings here, and extend them by examining the reasons
behind the lower self-concepts in certain areas. Why do
teenage boys with physical disabilities feel lower in scho-
lastiC competence? Once the reasons are. known, specific
interventions can be used, perhaps in a preventative way,
to address the issues. Clearly, not all adolescents with
physical disabilities are at risk for lower self-concepts.
However, if those who are at risk can be identified early,
before behavior patterns become more resistant to
change, then interventions may be easier and the burden
of concern may be alleViated. A
Acknowledgments
We thank Vicky Martin, Maria Gitta, and Sonya Vellet for their
assistance. We also thank]an Polgar and occu pational therapists
at Thames Valley Children's Centre for their input.
Preparation of this article was supported by a research
award from the Thames Valley Children's Centre and by a doc-
toral training grant, awarded to Kathleen Steel, from the Easter
Seal Research Institute.
References
Abramson, M., Ash, M. ]., & Nash, W. R. (1979). Handi-
capped adolescents -A time for reflection. Adolescence, 14,
557-565
Anderson, E. M., Clarke, L., & Spain, B. (1982). Disahility in
adolescence London: Methuen & Co.
Arnold, P., & Chapman, M. (1992). Self-esteem, aspirations,
and expectations of adoJescents with physical disability. Devel-
opmenlal Medicine and Child Neurology, 34, 97-102.
Blum, R. W., Resnick, M. D., Nelson, R., & St. Germaine, A.
139
 (1991). Family and peer issues among adolescents with spina
bifida and cerebral paiS)'. Pediatrics, 88, 280-285.
Brantley, H. T., & Clifford, E. (1979). Cognitive, self-
concerH, and body image measures of normal, cleft ralate, and
obese adolescents. Cleft Palate Journal, 16, 177-182.
Bryan, D. P., & Herjanic, B. (1980). Derression and suicide
among adolescents and young adults with selective handicap-
ping conditions. Exceptional Education Quarterly, 1, 57-65.
Campbell, M. M., Hayden, P. W., & Davenport, S. L. H.
(1977). Psychological adjustment of adolescents with myelodys-
plasia. Journal of Youth and Adolescence, 6, 397-407.
Canadian Association of Occupational Therapists. (1991a).
Occupational therapy gUidelines for client-centred practice.
Toronto: CAOT Publications.
Canadian Association of Occupational Therapists. (1991b).
Canadian occupational performance measure. Toronto:
CAOT Publications.
Cohen, J. (1988). Statistical power analysisfor the behav-
ioral sciences (2nd ed.). Hillsdale, N. ].: Erlbaum.
Connolly,]. (1989). Social self-efficacy in adolescence: Rela-
tions with self-concept, social adjustment and mental health.
Canadian Journal of Behavioral Science, 21, 258-269.
Crocker, ]., & Major, B. (1989). Social stigma and self-
esteem: The self-protective properties of stigma. Psychological
Review, 96, 608-630
Diener, E. (1984). Subjective well-being Psychological Bul-
letin, 95, 542-575.
Drotar, D. (1981). Psychological perspectives in chronic
childhood illness. Journal ofPediatric Psycholo[!J!, 6, 211-228.
Fitts, W. H. (1965). Manual for Tennessee Self-Concept
Scale. NashviJJe: Counselor Recordings and Tests.
Garmezy, N. (1983). Stressors of childhood. In N. Garmezy
& M. Rutter (Eds.), Stress, coping and development in children
(pp. 43-84). Minneapolis: McGraw-Hill.
Gordon, l. V. (1967). SunJey ofpersonal values. Chicago:
Science Research Associates.
Harper, D. C, & Richman, L C (1978). Personality profiles
of physically impaired adolescents. Journal ofClinical Psychol-
ogy, 34, 636-642.
Harter, S. (1978). Effectance motivation reconsidered: To-
ward a developmental model. Human Development, 21, 34-64.
Haner, S. (1986). Selfperception profile for adolescents.
Unpublished manuscript.
Kapp, K. (1979). Self concept of the cleft lip or ralate child.
Cleft Palate Journal, 16, 171-176.
Kapp-Simon, K. (1986). Self-concept of primary-school-age
children with cleft lip, cleft palate, or both. Cleft PalateJournal,
23,24-27
Kielhofner, G. (Ed.). (1985). A model of human occupa-
tion. Theory and application. Baltimore: Williams and Wilkins.
KeJJerman,]., Zeltzer, l., EJJenberg, L., Dash,]., & Rigler, D.
(1980). Psychological effects of illness in adolescence. 1. Anxi-
ety, self-esteem, and percertion of control. Journal of Pediat-
rics, 97, 126-13l.
Kirk, R E (1968). Experimental design: Proceduresfor the
behavioral sciences. Belmont, California: Brooks/Cole.
La Greca, A. M. (1990). Social consequences of rediatric
conditions: Fertile area for future investigation and interven-
tion? Journal of Pediatric Psychology, 15, 285-307.
Lord, ]., VarlOS, N., Behrman, B., Wicks, ]., & Wicks, D.
(1990). Implications of mainstream classrooms for adolescents
with spina bifida. Developmental Medicine and Child Neurol-
ogy, 32, 20-29.
Lorr, M., & Youniss, R. P. (1988). The interpersonal style
140
Downloaded From: http://ajot.aota.org/pdfaccess.ashx?url=/data/journals/ajot/930215/ on 02/27/2017 Terms of Use: http://AOTA.org/terms
inventory. Los Angeles: Western Psychological Services.
Magill, J, & Hurlbut, N. (1986). The self-esteem of adoles-
cents with cerebral palsy. Amencan Journal of Occupational
Therapy, 40, 402--407
Magill-Evans,]., & Restall, G. (1991). Self-esteem of persons
with cerebral palsy: From adolescence to adulthood. American
Journal of Occupational Therapy, 45, 819-825
Minde, K. K. (1978). Coping styles of 34 adolescents with
cerebral palsy. American Journal of Psychiatry, 135, 1344--
1349
Moos, R H., & Billings, A. G. (1982). ConceptualiZing and
measuring coping resources and processes. In l. Goldberger &
S. Breznitz (Eds.), Handbook ofstress: Theoretical and clinical
aspects (pp. 212-230). New York: Free Press.
Murch, R. I., & Cohen, l. H. (1989). Relationships among
life stress, perceived family environment, and the psychological
distress of spina bifida adolescents. Journal of Pediatnc Psy-
cholo[!J!, 14, 193-214
Pearson, A., Carr, ]., & HalliweJJ, M. (1985). The self-
concept of adolescents with srina bifida. Zeitschrijt fur Kinder-
cbirurgie, 40, 27-30.
Piers, E. V., & Harris, D. B. (1967). Piers-Harris Children's
Self-Concept Scale. Nashville: Counselor Recordings and Tests.
Pless, I. B., Crirrs, H. A., Davies,]. M. C, & Wadsworth, M.
E.]. (1989). Chronic physical illness in childhood: Psychological
and social effects in adolescence and adult life. Developmental
Medicine and Child Neurology, 31, 746-755.
Rae-Grant, N., Thomas, B. H., Offord, D. R., & Boyle, M. H.
(1989). Risk, protective factors, and the prevalence of behavior-
al and emotional disorders in children and adolescents.Journal
of the Amencan Academy ofChild and Adolescent Psychiatry,
28, 262-268.
Richman, I. C (1983). Self-reported social, speech, and
facial concerns and personality adjustment of adolescents with
cleft lir and palate. Cleft Palate Journal, 20, 108-112.
Rokeach, M. (1960). Tbe open and closed mind New York:
Basic.
Rutter, M. (1985). Resilience in the face of adversity. British
Journal qf Psychiatry, 147, 598--611
Shultz, 1. Z, Warr-Leeper, G., King, G. A., McGoldrick,
]. A., & Stanczyk, W. M. (1990). A conceptual model for social
skills inten;entions with neurologically and physically dis-
abled children and adolescents. Manuscript submitted for
publication.
Starr, P. (1978). Self-esteem and behavioral functioning of
teen-agers with oral-faCial clefts. Rehabilitation Literature, 39,
233-235
Strax, T. E. (1988). Psychological problems of disabled ado-
lescents and young adults. Pediatric Annals, 17, 756-76l.
Tavormina,]. B., Kastner, L. S., Slater, P. M., & Watt, S. L.
(1976). Chronically ill children: A psychologicaJly and emotion-
ally deviant population? Journal ofAbnonnal Child Psychology,
4, 99-110.
Taylor, S. E., & Brown, J (1988). Illusion and well-being:
Some social psychological contributions to a theory of mental
health. Psychological Bulletin, 103, 193-210.
Werner, E. E., & Smith, R. S. (1982). Vulnerable but invin-
cible: A longitudinal study of resilient children and youth.
New York: McGraw-HilI.
Wright, B. (1960) Pbysical disability: A psychological ap-
proach. New York: Harper & Row.
Yuker, H. E., Block, J. R., & Younng, ]. H. (1966). The
measurement ofattitudes toward disabled persons. Albertson,
NY: Human Resources Center.
February 1993, Volume 47, Number 2