By Irmo Marini, Lee Rogers, John R. Slate and Cheryl Vines

Self-esteem may be one major factor related to the manner in which people
with SCI respond to rehabilitation. Self-esteem, believed to be a major
factor of our self-concept, may be considered as an individual's global
positive or negative feelings toward himself or herself (Rosenberg, 1979). A
person's self-esteem also affects the manner in which he or she deals with
the environment. That is, people with low self-esteem tend to view their
environment as threatening, and, therefore, have difficulty interacting in
it (Roy, 1976). People with high self-esteem, however, tend to deal more
actively with their environment, are better able to meet environmental
demands, and generally feel more secure about themselves (Coopersmith, 1967;
Zejdlik, 1992).

Battle (1992) indicated that self-esteem is a construct involving a person's
perception about his or her own worth. Although initially not well
developed, self-esteem is believed to evolve as people mature, resulting
from interactions with significant others and various personal life
experiences. Branden (1980) stated that when it is developed, self-esteem
remains fairly stable and resistant to change unless an individual goes
through an extended period of evading decisions, consistently failing, and
no longer being productive. Self-esteem, therefore, must be maintained
through continual successful interactions with one's environment.

Although it is commonly assumed that the person with acute SCI will
experience serious psychological disruption, many researchers disagree with
this assumption (Frank et al., 1987; Gunther, 1971; Livneh, 1991a). Their
disagreement stems from the scarcity of empirical data supporting the loss
of self-esteem among people with SCI. The purpose of conducting this study
was to address the lack of data regarding self-esteem of people with SCI.
Specific research hypotheses in this ex post facto study were as follows:

1. People who sustain a SCI perceive a loss of self-esteem.

2. Perceived levels of self-esteem increase over time.

3. Demographic variables are related to self-esteem.

METHOD

Participants

There were 63 participants for this study (49 men, 14 women) with SCIs who
were registered with the Arkansas Spinal Cord Commission (ASCC) and living
in Arkansas. In Arkansas, the ASCC is contacted by hospital personnel when
someone is admitted with a spinal injury. The individual is then registered
and a case manager from the ASCC is assigned. Ethnicity of participants was
Caucasian, 71% (n = 45); African American, 19% (n = 12); Native American, 8%
(n = 5); and Hispanic, 2% (n = 1). Participants' ages ranged from 16 to 24
(21%), 25 to 34 (35%), and 35 to 44 (22%). Levels of SCI were paraplegia
47.5% (n = 30), quadriplegia 43% (n = 27), and not reported 9.5% (n = 6).

Regarding employment before the injury, 76% (n = 48) of the participants
indicated that they were employed, 9.5% (n = 6) indicated that they were
unemployed, and 14.5% (n = 9) did not provide this information. Following
SCI, only 9.5% of the sample reported being employed, 46% (n = 29) indicated
that they were unemployed, 17.5% (n = 11) listed themselves as retired, 14%
(n = 9) indicated student status, 5% (n = 3) stated homemaker status, and 8%
(n = 4) did not provide this information.

Responses on the highest level of education completed were graduate degree
3% (n = 2), bachelor's degree 6% (n = 4), some college 21% (n = 13), high
school diploma 33% (n = 21), some high school 32% (n = 20) and 5% (n = 3)
that did not provide this information.

Participants' living arrangements were as follows: 40% (n = 25) in a house,
19% (n = 12) in an apartment, 29% (n = 18) living with parents, and 19% (n =
12) living with a spouse. Of the respondents, 46% (n = 29) were married, 21%
(n = 13) were single and dating, 19% were single and not dating, and 8% (n =
5) were divorced following SCI, whereas 6% were divorced before their injury
occurred.

When asked about the manner in which their injury occurred, 30% (n = 19) of
our sample reported being in a motor vehicle accident in which they were the
driver; an additional 13% (n = 8) were in an automobile; 17% (n = 11)
experienced a diving injury or fall; and 35% (n = 22) either attributed
their injury to other causes, or did not respond to this question When asked
whether their accident could have been avoided, 36.5% (n = 23) of
participants said no, 16% (n = 10) responded yes, 30% (n = 19) did not know,
and 17.5% (n = 11) did not respond.

Of the respondents, 41% (n = 26) reported their SCI had occurred 5 years
previously (1986 to 1987), 36.5% (n = 23) reported their injury occurred
approximately 2 years previously (1990 to 1991), and 22% (n = 14) reported
having sustained SCI in the previous year (1992 to 1993). Health status was
perceived as good (30%), average (27%), excellent (9.5%), poor (6%), or very
poor (6%), with 17% not providing this information.

Procedure

All 150 people with SCI injured during 1986 to 1987, 1990 to 1991, and 1992
to 1993 who were registered with the ASCC were mailed a survey that included
an informed consent form, an invitation and explanation of the study letter,
an instruction sheet for completing the -inventories, a demographic
information form, and two identical copies of Form AD of the Culture-Free
Self-Esteem Inventory-2 (CFSEI-2). The only difference in each of the two
CFSEI-2s was an instruction in large print. One read "How I felt before my
injury" and the other read "How I feel today." In cases where solicited
persons had not responded by the deadline, a second mailing was sent 2 weeks
later. Although a response rate of 48% was obtained, only 42% (N = 63) of
the forms were completed sufficiently to be analyzed in this study.

Instrumentation

The Culture-Free Self-Esteem Inventory (Battle, 1981; 1992) is designed to
measure a person's perception of worth or self-esteem. It was selected for
its ability to break down self-esteem into separate constructs as discussed
in the literature on self-esteem (Battle, 1992; Branden, 1980). As Battle
(1992) indicated, although defining the concept of self-esteem is somewhat
controversial among researchers, it is generally agreed that self-esteem is
composed of separate constructs.

Our study involved using the adult version comprising 40 yes-no response
type questions (Form AD) producing a total score consisting of four separate
subscales: general, social, personal, and lie. The general subscale refers
to an individual's overall perceptions of worth. The social subscale refers
to perceptions of relationships with friends. The personal subscale refers
to an individual's intimate perceptions of his or her own self-worth.
Finally, there is a lie subscale that measures defensiveness and contains
items related to matters considered socially undesirable. The test authors
reported concurrent validity ranging from .71 to .80 when correlated with
Coopersmith's (1967) Self-Esteem Inventory (Battle, 1992). In the current
study, the internal consistency (i.e., coefficient alpha) of the total
CFSEI-2 test was .76, which compares favorably to the reliability estimates
reported in the test manual (Battle, 1992). Additional internal consistency
estimates of the four subscales, pre- and postinjury, produced average
alphas of .85 for the general subscale, .69 social subscale, .81 personal
subscale, and .57 for the lie scale.

Data Analysis

To ascertain whether overall differences in self-esteem were present pre-and
postinjury, a t test was conducted. Analyses of variance were performed on
the subscales of the CFSEI-2 to determine what differences, if any, were
present in self-esteem as a result of year since injury. Finally, analyses
of variance were conducted to calculate differences in self-esteem with 15
demographic variables using the Bonferroni procedure.

RESULTS

The variable measured in this study was self-esteem scores derived using the
CFSEI-2. The manipulated time variables were the three periods of time since
injury. In response to the first research hypothesis, that people who
sustain SCI perceive a loss of self-esteem, perceived self-esteem was
significantly higher before SCI (M = 28.29) than after SCI (M = 21.79),
t(60) = 4.58, p < .01 on the overall scale. Moreover, an analysis of
variance revealed a statistically significant difference between years since
injury and perceived self-esteem, F(2,59) = 4.00, p < .05. That is,
differences were present between postinjury total self-esteem scores,
F(2,59) = 5.53, p < .01, but not between self-esteem level before injury
F(2,59) = 1.68, p > .05. The 2-year postinjury group reported a
significantly lower self-esteem (M = 17.50) than did the 5-year postinjury
group (M = 25.75), F(1,49) = 5.57, p < .05, which scored highest among the
three groups.

Next, analysis of variance procedures were conducted on the three subscales
of the CFSEI-2 for the postinjury groups. Statistically significant
differences between groups were noted on all three subscales: general
self-esteem, F(2,59) = 5.85, social self-esteem, F(2,59) = 4.92, and
personal self-esteem F(2,59) = 8.82, all ps < .01. Follow-up Scheffe tests
on the general subscale indicated a significant difference between the
2-year postinjury group (M = 6.32) and the 5-year postinjury group (M =
10.38). Next, Scheffe comparisons with the social subscale indicated that
the 2-year postinjury group scored significantly lower in social self-esteem
(M = 4.45) than did the 1-year group (M = 6.14) and the 5-year group (M =
5.87). The social subscale was the only scale in which the 1-year group
scored highest. Follow-up Scheffe tests on the personal subscale indicated
that the 5-year postinjury group scored significantly higher (M = 5.0) than
did the 2-year postinjury group (M = 1.85), but not the l:year group. Thus,
in response to the second research hypothesis, self-esteem does increase
over time following injury. Perceived social self-esteem seems greatest
during the first year, however. See Table 1 for means and standard
deviations as well as classifications of scores (i.e., low, moderate, high)
according to Battle's (1992) normative sample.

Finally, in response to the third research hypothesis, analyses of variance
were performed on 15 demographic factors and self-esteem. To control for
experimentwise Type I error (Parker & Szymanski, 1992), the Bonferroni
procedure was applied, revealing no statistically significant differences
between perceived total self-esteem and the variables tested.

DISCUSSION

Results from this study both support and modify previous assumptions
regarding perceived self-esteem changes following SCI. Consistent with
earlier findings (e.g., Nelson, 1987; Piazza et al., 1991; Roy, 1976;
Trieschmann, 1988), the current sample did report a perceived lower
self-esteem following SCI. Although not statistically supported in this
study, this finding may be partially explained by the large drop in
employment experienced by people with SCI following their injury. Before
their injuries, 76% of the group were employed, compared to the 9.5%
employed after injury. Statistics from the ASCC registry (1985 to 1992)
indicated that more than 33% of people with SCI were employed in a
physically demanding occupation (e.g., operators, laborers) before their
injury, which likely made it difficult to return to their former jobs (ASCC,
1993). Because of the significance and value that Western society places on
work (Power, 1991), the loss of employment may further exacerbate the
already lowered self-esteem of persons with SCI.

Also consistent with previous findings (Cook, 1979; Livneh, 1991b; Nelson,
1987; Trieschmann, 1988) was that the longer the time period since SCI, the
higher the perceived self-esteem. That is, the 5-year postinjury group in
this study scored the highest in how they perceived personal self-esteem.
Despite external circumstances, people with SCI seem to adjust and
reestablish a new role over time.

A final consistency with previous findings reflects the lack of perceived
self-esteem differences regarding severity of SCI. Fuhrer, Rintala, Hart,
Clearman, and Young (1992) stated, "From the standpoint of an augmented
version of the World Health Organization model of disablement, the life
satisfaction of persons with spinal cord injury appears to be influenced,
albeit indirectly, by selective aspects of their social role performance
(disablement), but not by their degree of impairment or disability" (p.
552).

In disagreement with existing research (although participants' perceived
their self-esteem did indeed decrease following their SCI), even lower
levels of self-esteem were reported among the group who were in their second
year of injury. This finding does not support the assumption that
self-esteem is most vulnerable initially following SCI. Rather, the current
study data suggest that self-esteem may be most vulnerable in the second
year of injury. Several explanations for this finding are plausible. First,
initially following injury, the injured person is surrounded with social
support from friends and family. This support often continues during the
ensuing months of hospitalization. Second, people with SCI may, during
hospitalization, perceive themselves as sick, thus validating a lack of
activity, absence of work status, and other societal pressures generally
relinquished for people who are ill (Wright, 1983). Finally, the attitudes
of the staff at rehabilitation hospitals may be more accepting of people
with SCI than is society in general (Makas, 1988). Two years postinjury,
however, most people with SCI have returned home, made maximum physical
strength gains, are unemployed, and experience a markedly lower o level of
social support than during hospitalization. In essence, although the person
with SCI initially has a valid reason for his or her role or circumstances
(e.g., "I am currently sick, therefore I am temporarily excused of my former
duties") 2 years after injury, and now stabilized and at home, the injured
person may perceive him- or herself as "roleless," or discontent with the
"sick" role society assumes for persons with disabilities (Wright, 1983).

Variables such as age, marital status, living arrangements, and educational
level did not support the findings of other researchers (Craig, Hancock,
Dickson, Martin, & Chang, 1990; Frank, Elliott, Buckelew, & Haut, 1988;
Mayer & Andrews, 1981; Schulz & Decker, 1985;). Failure to support existing
research may have occurred because previous studies explored factors
believed to be associated with successful adjustment (e.g., assertiveness)
rather than self-esteem.

Readers should be aware of several limitations in this research study.
First, the current sample of participants was from a geographically
restricted location, one that is primarily rural. If environmental factors
such as transportation influence self-esteem level (e.g., through an
inability to visit a friend), these factors deserve consideration. Second,
usable data were obtained on only 42% of the initial pool of people with
SCI. Because self-selection can produce skewed results, the extent to which
these findings generalize to the population of people with SCI is unknown.
In addition, no demographic information was reviewed regarding
characteristics of people who did not respond to the study. Third, a
retrospective methodology in assessing participants' self-esteem was used
here. The request that participants recall their self-esteem several years
previously as well as indicate their current self-esteem could have
unknowingly introduced bias (i.e., selective memory), possibly resulting in
inflated preinjury estimates of self-esteem. Finally, we found no previous
studies measuring perceived self-esteem at different periods. Individuals
may have idealized that their self-esteem was higher before their injury,
when it actually may not have been higher. Obviously, longitudinal research
with people with SCI is needed to ascertain the viability of the study
design used here.

In summary, these findings suggest that the second year following SCI may be
just as critical a time period as the initial one after injury, pertaining
to vulnerabilities in self-esteem. If replicated, this finding is of
potential importance in the rehabilitation field. For example, the majority
of psychological services are currently available at the acute stage of
injury. People with SCI are most often released from the hospital when they
are physiologically stable and physically deemed ready for release. It is
precisely at this point, however, that people with new SCI may have many
questions and concerns regarding what their new role will be in the
community. The current study suggests that perhaps practitioners need to
make a more concerted effort in dealing with the psychological effects of
SCI when difficulties may really begin, out in the community.

TABLE 1 Means and Standard Deviations of the CFSEI-2 Total and Subscale
Scores by Years Since Injury
Legend for Chart:

A - Years Since Injury
B - n
C - Total Score, M
D - Total Score, SD
E - Personal, M
F - Personal, SD
G - Social, M
H - Social, SD
I - General, M
J - General, SD

A B C D E
F G H I J

5 years 26 25.75[a] 8.26 5.00[a]
2.73 5.87[a] 1.77 10.38[a] 4.12

2 years 23 17.50[b] 6.62 1.85[c]
1.63 4.45[a] 1.89 6.32[b] 3.69

1 year 14 22.36[a] 8.70 3.86[b]
3.13 6.14[d] 1.79 7.45[a] 4.18

Note. CFSEI-2 = Culture-Free Self-Esteem Inventory, 2nd edition. The values
represent group mean and standard deviation scores for the total inventory
and its three subscales.

a Classified as intermediate self-esteem.

b Classified as low self-esteem.

c Classified as very low self-esteem.

d Classified as high self-esteem.

REFERENCES

Arkansas Spinal Cord ComÅÇÉÑÖÜáàâäãåçéèêë'""ïñóòôöõúùûü+ocPound$*Pß®©(TM)'"<>ÆØ(inf)+/-<=>=YenmdSPpS(a)(o)Oæø?!!sqrt f~=D<<>>... ÀÃÕOEoe-",^^^,^^^'~^^^/#ÿÿ/o<>fifl++.'"o/ooÂÊÁËÈ'"Ï"ÓÔ@ÒÚÛÙi^~-^.(o),~,^mission. (1993). Arkansas spinal cord disability
registry [Computer program]. Little Rock, AR: Author.

Battle, J. (1981). Culture-Free Self-esteem Inventories for children and
adults. Seattle, WA: Special Child Publications.

Battle, J. (1992). Culture-Free Self-esteem Inventories (2nd ed.). Austin,
TX: Pro-Ed.

Branden, N. (1980). A Nathaniel Branden anthology. Los Angeles, CA: Tarcher.

Cook, D. (1979). Psychological adjustment to spinal cord injury: Incidence
of denial, depression, and anxiety. Rehabilitation Psychology, 26, 97-104.

Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco, CA:
Freeman.

Craig, A., Hancock, K., Dickson, H., Martin, J., & Chang, E. (1990).
Psychological consequences of spinal injury: A review of the literature.
Australian and New Zealand Journal of Psychology, 24, 418-425.

Frank, R., Elliott, T., Buckelew, S., & Haut, A. (1988). Age as a factor in
response to spinal cord injury. American Journal of Physical Medicine and
Rehabilitation, 67(3), 128-131.

Frank, R., Ulmlauf, R., Wonderlinch, S., Askunazi, G., Bechela, S., &
Elliott, T. (1987). Differences in coping stages among persons with spinal
cord injury: A cluster-analytic approach. Journal of Consulting and Clinical
Psychology, 55(5), 727-731.

Fuhrer, M., Rintala, D., Hart, K., Clearman, R., & Young, E. (1992).
Relationships of life satisfaction to impairment, disability, and handicap
among persons with spinal cord injury living in the community. Archives of
Physical Medicine and Rehabilitation, 73, 552-557.

Gunther, M. S. (1971). Psychiatric consultation in a rehabilitation
hospital: A regression hypothesis. Comprehensive Psychiatry, 12(6), 572-585.

Koehler, M. (1989). Relationship between self concept and successful
rehabilitation. Rehabilitation Nursing, 14(1), 9-12.

Livneh, H. (1991a). On the origins of negative attitudes toward people with
disabilities. In R. P. Marinelli & A. E. Dell Orto (Eds.), The psychological
and social impact of disability (3rd ed., pp. 181-196). New York: Springer.

Livneh, H. (1991b). A unified approach to existing models of adaptation to
disability: A model of adaptation. In R. P. Marinelli & A. E. Dell Orto
(Eds.), The psychological and social impact of disability (3rd ed., pp.
181-196). New York: Springer.

Makas, E. (1988). Positive attitudes toward disabled people: Disabled and
nondisabled persons' perspectives. Journal of Social Issues, 44(1), 49-61.

Mayer, T., & Andrews, H. (1981). Changes in self-concept following a spinal
cord injury. Journal of Applied Rehabilitation Counseling, 12(3), 135-137.

Nelson, A. (1987). Normalization: The key to integrating the spinal
cord-injured into the community. SCI Nursing, 4, 3-6.

Parker, R. M., & Szymanski, E. M. (1992). Fishing and error rate problem.
Rehabilitation Counseling Bulletin, 36(2), 66-69.

Piazza, D., Holcombe, J., Foote, A., Paul, P., Love, S., & Duffin, P.
(1991). Hope, social support and self-esteem of patients with spinal cord
injuries. American Association of Neuroscience Nurses, 23(4), 224-230.

Power, P. W. (1991). A guide to vocational assessment (2nd ed.). Austin, TX:
Pro-Ed. Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.

Roy, C. (1976). Introduction to nursing: Adaptation model. Englewood Cliffs,
NJ: Prentice-Hall.

Schulz, R., & Decker, S. (1985). Long-term adjustment to physical
disability: The role of social support, perceived control, and self-blame.
Journal of Personality and Social Psychology, 48(5), .

Stover, S., & Fine, P. (1986). Spinal cord injuries: The facts and figures.
Birmingham, AL: National Spinal Cord Injury Research Data Center, University
of Alabama.

Trieschmann, R. (1988). Spinal cord injuries (2nd ed.). New York: Demos
Publications.

Wright, B. A., (1983). Physical disability: A psychosocial approach. New
York: Harper & Row.

Zejdlik, C. (1992). Management of spinal cord injury (2nd ed.). Boston, MA:
Jones and Bartlett.

~~~~~~~~

By Irmo Marini, Lee Rogers, John R. Slate and Cheryl Vines
Irmo Marini is an assistant professor and John R. Slate is a professor, both
in the Counselor Education and Psychology Department, Arkansas State
University. Lee Rogers is a mental health worker at George W. Jackson Mental
Health Center, Joneboro, Arkansas. Cheryl Vines is the executive director of
the Arkansas Spinal Cord Commission, Little Rock. The authors would like to
extend their appreciation to Constance Carroll, Tom Farley, and Dee
Ledbetter of the Arkansas Spinal Cord Commission (ASCC) for conducting the
mailing and follow-up of survey materials, as well as their expertise in
codereloping a demographic form for identifying persons within the ASCC
registry. Correspondence regarding this article should be sent to Irmo
Marini, Counselor Education and Psychology Department, Box 1560, State
University, AR 72467.

Copyright 1995 by American Counseling Association. Text may not be copied without the express written permission of American Counseling Association.
Marini, et, Self-esteem differences among persons with spinal cord
injury.., Vol. 38, Rehabilitation Counseling Bulletin, 03-01-1995, pp 198.
===============================================================