Graduate Student Paper
| Professor
comments: |
Graduate Student Paper:
Reviewing the Life Satisfaction in the Elderly Scale
(LSES)
|
| |
Abstract
This study reviewed the Life Satisfaction in the Elderly Scale (LSES)
that was developed by Salamon and Conte in 1984. The LSES was designed
in order to produce an assessment tool with improved psychometric
properties that could be useful in measuring an elderly (55 and older)
individual's self-reported perception of life satisfaction. The findings
of this review indicate that the LSES has stronger reliability and
validity levels than previously developed scales, and measures the
domains of Pleasure in daily activity, Meaningfulness of life, Goodness
of fit between desired and achieved goals, Mood tone, Self-concept,
Social contact, Perceived health, and Financial security. All are
considered important characteristics of the abstract construct of life
satisfaction. The LSES has been used in numerous research studies and is
considered useful in clinical assessments of depression, and as a
rapport building tool during the counseling session. |
| |
Beginning with Erik Erikson's developmental theory and the
concept of "ego integrity versus despair in late adulthood,"
more than 50 years of research and assessment attempts have been
introduced in an effort to appraise the elderly's satisfaction with life
and their sense of well-being (Barrow, 1996; Phillips, 1986; Rathus,
1993). Erikson purported that the aging person is faced with the
challenge of retaining their individual belief that life remains
meaningful in view of approaching death (Euler, 1992; Rathus, 1993). It
is this underlying theoretical concept and the hope that once
satisfaction in life is defined, advances could be made to assist the
aging populace to further achieve a comfortable level of contentment in
the final years (Conte & Salamon, 1982; Phillips, 1986; Ryff, 1995;
Schultz & Heckhausen, 1996). This perspective has repeatedly stirred
the social and gerontological sciences to labor to quantify and qualify
the constructs of satisfaction and subjective well-being
(Barrow, 1996; Neugarten, Havighurst & Tobin, 1961; Phillips, 1986;
Salamon, 1988; Conte & Salamon, 1982). |
| |
Salamon and Conte (1982, 1988) have operationally defined life
satisfaction as the perceived degree to which an individual has
successfully cleared life's developmental circumstances. These
developmental circumstances are things such as overcoming childhood
illnesses, the process of education, to marry or to remain single,
employment and/or job loss, and coping with the loss or death of loved
ones (e.g., through divorce or runaway, etc.)(Conte & Salamon,
1982). |
| Excellent introduction: strong
foundation and definition of construct |
Well-being is further defined as an individual's
perceived state of health, prosperity, happiness, sense of
accomplishment, and comfort in life (Ryff, 1995). In addition, elderly
is defined as an individual aged 55 and older (Conte & Salamon,
1982). This evaluation of life satisfaction is subjective in that
it is one's personal opinion of their adaptation processes and their
general well-being (Conte & Salamon, 1982). Neugarten,
Havinghurst and Tobin (1961) advocated that only through the use of a
self-report instrument could the sciences appropriately measure a
person's subjective well-being; after all, only the individual is the
best judge of their own happiness and contentment with life (p. 134). |
| |
Salamon and Conte authored the Salamon-Conte Life
Satisfaction in the Elderly Scale (SCLSES) in 1984 (title later changed
to LSES; Salamon, 1988), in order to develop an assessment instrument
with sound psychometric properties that would improve on the then widely
used Life Satisfaction Index (LSI) scale developed by Neugarten, et al.
(1961) (Conoley, Kramer & Mitchell, 1988; Conte & Salamon,1982;
Salamon, 1988). Aside from improving on the psychometric properties of
the LSI, Salamon and Conte also expanded the coverage of areas on the
instrument to include other facets that elderly individuals considered
important at their particular stage of life. This additional information
was obtained through extensive literary review and survey of the elderly
(Brockett, 1987; Conoley, et al, 1988; Gould, 1992; Neri, Andermarcher,
Pradelli & Salvioli, 1995; Salamon, 1988). |
| |
Test Description
The LSES, a multi-factor scale, is constructed of 40 multiple choice
questions believed to measure the construct of well-being (Salamon,
1988; Conte, Salamon, 1982). Five of the eight subscale domains are
taken from the LSI scale. Neugarten, et al. (1961), and later Salamon
and Conte (1984) and others (Conoley et al, 1988; Conte & Salamon,
1982; Phillips, 1986) have hypothesized that these constructs are
measures of psychological well-being, happiness, morale, and/or
contentment in life (terms used interchangeably throughout the literary
review). |
| |
The five basic domains of satisfaction measurement
incorporated in the LSES and borrowed from the LSI are: Pleasure in
daily activities, Regarding life as meaningful, Desired versus achieved
goals, Positive mood, and Positive self-concept (Conoley et al, 1988).
The other three domains that were added later are: Perceived health,
Perceived financial security and Satisfaction with social contacts (Conoley
et al, 1988). The three that were later added by Salamon and Conte, were
based on Spreitzer and Snyder's (1974) research that indicates that the
perceptions of finances, health and social involvement were influential
considerations to the elderly, and must be factored into the overall
concept of life satisfaction (Conoley et al, 1988; Conte & Salamon,
1982; Salamon, 1985, 1988; Gould, 1992). |
| |
Conte and Salamon (1982) advise that the different domains
are generally written as open-ended sentences and are structured to
measure different concepts of the theoretical description of life
satisfaction. These domains, a brief description, an example of an item
within that subscale, and the Likert choice of answers are as follows:
1. Pleasure in daily activity: measures the individual's
degree of satisfaction with numerous, unspecified daily activities of
the older person. (i.e., My daily routine is: very boring, boring,
average, satisfying, very satisfying)
2. Meaningfulness of life: reflects the individual's positive
attitude toward life as a result of a sense of usefulness and purpose in
his or her life situation. (i.e., I am most satisfied with my life
situation: never, almost never, sometimes, often, always)
3. Goodness of fit between desired and achieved goals:
involves matching the individual's degree of satisfaction with his or
her present stage of life relative to previous stages. (i.e., In my life
I have achieved: nothing, very little, something, a lot, a great deal)
4. Mood tone: This subscale allows for a measure of general
positive effect, happiness, or optimism, nonspecific with regard to
one's environment or social situation. (i.e., I am ___ in a bad mood:
always, often, sometimes, seldom, never)
5. Self-concept: Measures the degree of personal self-regard
and favorable self- appraisal. (i.e., Compared to any other time in my
life, I am now: very unsatisfied, unsatisfied, average, satisfied, very
satisfied)
6. Perceived health: Measures the individual's self assessment
of overall physical well-being. (i.e., Physically I am: unhealthy,
somewhat unhealthy, average, healthy, very healthy)
7. Financial security: The individual evaluates their own
present and recent financial situation. (i.e., I have enough money to
enjoy myself: never, rarely, sometimes, often, always)
8. Social Contact: Measures one's perceived level of
satisfaction with the number and quality of social contacts that are
characteristic of the respondent's usual routine. (i.e., I try to spend
time with people: never, rarely, sometimes, often, always) (p. 197) |
|
Use 'e.g.,' instead of 'such as'
|
Each test kit, published by Psychological Assessment
Resources, Inc., costs $43 and includes 50 scoring sheets, 50 test
booklets and the 27 page test manual that was revised in 1992 (M. J.
Salamon, personal communication, April 1999). Items are written at a
sixth grade reading level and are addressed in the first-person (Conoley,
Kramer & Mitchell, 1988; Phillips, 1986). The authors advise that
this pencil and paper instrument is easy to administer and score, using
simple addition. No special training is required to administer and/or
score. The LSES takes approximately 20 minutes to complete, and is
considered non-threatening by the test taker (Conte & Salamon,
1982). They further advise that in the event of physical or visual
impairment (such as arthritis, poor eyesight and/or illiteracy), another
person could read the items to the elderly individual and score their
responses on the score sheet provided. Instructions are included in the
test manual regarding scoring and possible interpretations (Conoley et
al, 1988; Conte & Salamon, 1982). |
| Highest value? indicate
valance
Clarify that items are summed.
|
Test Scoring
The LSES is constructed as a self-report instrument in a five-point
Likert scale format. The scores for each item vary from one to five,
with five being the highest value possible (Conoley et al, 1988; Conte
& Salamon, 1982; Phillips, 1988). The test-taker merely places an
"X" on the line above one of the five responses provided for
each item; each response yields a numerical score of 1 to 5. This
instrument produces a total score of 40 to 200 points, and subscale
scores range from 5 to 20. The analysis is that the higher the score,
the higher the individual's assessment of their satisfaction with life (Conoley
et al, 1988; Phillips, 1986; Salamon, 1987, 1988). |
| What other normative factors should
be considered?
Is this sample large enough?
|
Normative Sample
The LSES was administered in three different studies to 649
individuals who lived in the New York Metropolitan area (Conoley et al,
1988; Salamon, 1984, 1988). Conte and Salamon (1982) report that
subjects were randomly solicited from various health care facilities,
nursing homes, assisted living quarters, and retirement communities in
order to access a broad spectrum of the socioeconomic levels. The
authors advise that the normative sample was based on data from the 1980
census and represents the various socio-economic levels (Conte &
Salamon; 1982; Salamon, 1988). All participants were 55 to 90 years of
age (Conoley et al, 1988; Conte & Salamon, 1982). |
| |
Reliability
Test-Retest
Reliability is the term used to define the dependability and/or
consistency of a test when used under similar circumstances (Cohen,
Swerdlik & Phillips, 1996). Information regarding the reliability
estimate of the LSES is confusing. Dixon reports test-retest reliability
coefficients estimates of .90 at one month time and .67 after a 6 month
time lapse (Conoley et al., 1988). Dixon did not specify in his
evaluation whether these coefficient estimates are for the total 649
individuals or only a subset of the sample. Neither did he indicate if
the test-retest values were based on subscales or total scores, however,
he did comment that "subscale reliabilities are adequate for most
subscales; however, the Goals and Self-concept subscales are
questionable due to the subjective dependence on present happiness
(coefficient alpha reliabilities of .60, .50 and .61, .47 respectively
for two different samples)" (Conoley et al, 1988, p.170). |
|
Add spaces to (n=50) to read (n = 50)
|
Salamon (1988) indicated that at the 6 month test-retest
interval, with 57of the original sample, a reliability coefficient of
.67 was measured based on total scores. Busch-Rossnagel reported a
test-retest reliability estimate on 120 of the original sample, of .67
for the total score; while at the one month test-retest interval, a
total score estimate of .90 (n=50) and reliability coefficient estimate
above .88 for all subscales was measured for fifty of the sample group (Conoley
et al., 1988, p. 172). |
| |
Salamon (1988) commented that these scores indicate a
"high moderate range" for test-retest reliability (p. 48).
This appears to be a liberal interpretation of conflicting data.
However, assuming the test-retest reliability estimates of .90 and .67
are accurate figures for one month and 6 month intervals, one could
surmise that the LSES reflects acceptable reliability and is sensitive
to change over time within the elderly population (Conoley et al, 1988). |
| |
Internal Consistency
Internal consistency computations, based on the total scale,
produced high coefficient alphas of .93 and .92 on two sample groups (Conoley
et al, 1988). Conte and Salamon (1982) report that after computing
Spearman Brown coefficients for the individual scores of each of the
initial five scales, the average coefficient was .95, which indicates
highly significant reliability. Busch-Rossnagel reports that moderate
subscale reliabilities range from .47 to .79 for two different sample
groups; with the subscales of Desired versus achieved goals and Positive
self-concept ranking the lowest with moderate reliability coefficient
scores of .47 to .61 (Conoley et al, 1988). |
| Check spelling of 'Crombach'
Delete 'to item'
Add spaces to 'p<.05' to read 'p < .05'
|
Brockett (1987) used the LSES in his study of elderly
people and their self-reported level of education. He reports a Crombach
alpha score of .88, which he interpreted as a high degree of consistency
(Brockett, 1987; Salamon, 1988). Brockett (1987) then conducted an item
to item analysis of the scale and compared it to the total scale and
found that 37 of the 40 items correlated strongly at 92.5%, p<
.05,
which adds additional support of internal consistency of this
instrument; which simply means that the items in the scale are measuring
the same thing (p. 232-233). |
| |
Validity
Construct Validity
Construct Validity has been defined as the overarching type of test
validation, in that it is based on the definition of the construct being
measured within context, time and place (M. McCarthy personal
communication, September 2, 1999). As stated earlier, the authors define
"life satisfaction . . . as the individual's assessment of their
progress toward perceived life's goals" (Salamon, 1988; p. 48).
Conte and Salamon hypothesized that the higher the total test score, the
higher the measure of life satisfaction of the individual (Brockett,
1987; Conte & Salamon, 1982; Salamon, 1987). |
| |
Factor analysis and cluster analysis were performed on the
eight factors. Salamon (1987,1988) reports that the perceived level of
health and financial security both correlated highly, while there was
moderate correlation among the other six categories. |
Interpret!
A summary statement needs to be here. |
Further reports indicate that a sophisticated rotation
analysis was performed on the eight subscales, and four of the five
items from five subscales showed high homogeneity (Conoley et al.,
1988). Salamon (1988) stated that these items correlated in the
hypothesized positive manner; such as a high perception of mood and
health would strongly correlate with a high measure of daily activities
and/or social contact and thus, a perceived high level of life
satisfaction (p. 48). The inverse would also correlate equally (Salamon,
1987, 1988). |
| |
Content Validity
Busch-Rossnagel reported that information in the manual confirms the
face validity of this instrument was based on the evaluation of four
experts, but that no additional information was provided (Conoley et
al., 1988, p. 171). A current examination of literary articles
pertaining to the LSES, also yielded no information to confirm or
discredit this statement; nor was there any recorded appraisal of face
validity by the test takers. |
| |
Criterion Validity
Dixon and Busch-Rossnagel stated that no criterion validity was
reported in the 1984 manual (Conoley et al., 1988). Salamon (1988) later
advised that scores from the second sample group, (n = 241) correlated
at an above average range of r = .85 when compared to individual medical
records from the local health care facility and social service intake
reports (p. 49). |
| Interpret why this is criterion
validity |
Salamon (1988) advised that a separate study involving 20
elderly living in a long term health care facility was conducted, using
the LSES, the Beck Depression Inventory (BDI), and the Geriatric
Depression Scale (GDS). These instruments were administered to this
small sample group. Statistical analysis revealed that total LSES and
BDI correlated at r = .48, and the GDS at r = .51. The mood subscale of
the LSES correlated moderately with BDI at r = .73 and with the GDS at r
= .78. Salamon (1988) considered these correlations to be significant,
yet low to moderate (p. 49). |
| Interpret |
In another study, Salamon (1988) reported that there was a
strong correlation between the LSES and other measures of health and
well-being. He referred to a study of 75 applicants to a health care
clinic for the elderly. These individuals completed the LSES, Health
Background Questionnaire and were also extensively interviewed during
intake procedures. Accordingly, Salamon (1988) stated that "highly
significant correlations resulted p> .00l between the LSES and the
additional data collected" (p. 49). |
| |
Brockett (1987) also used the LSES in his study with 64
elderly in which he was striving to determine if the individuals
self-directed learning experiences correlated with their perceived life
satisfaction. The subjects in this study were asked to complete the LSES
and the Self-Directed Learning Readiness Scale (SDLRS). The SDLRS is a
58-item Likert scale designed to measure an individual's perception of
their skills and attitudes as self-directed learners (Brockett, 1987; p.
230). His study revealed that five of the eight subscales correlated at
.21 to .24 using a Product Moment Correlation Coefficient with p< .05
with the SDLRS. These results were interpreted by Brockett as low, yet
significant and indicate that the individuals that retain a strong sense
of independence and control also self-report higher levels of
satisfaction with life (p. 231-233). |
| How is this criterion validity? |
The LSES was also used in additional studies measuring
such things as the individual's involvement in social groups and their
level of life satisfaction (Salamon, 1985); the perception of life
satisfaction while living in an long term care institution (nursing
home) (Gould, 1992; Salamon, 1987); the level of androgyny exhibited in
the elder as they progress through the aging process and their perceived
level of life satisfaction (Dean-Church & Gilroy, 1993); and as a
pre-test and post-test assessment tool used by clinical counselors after
intervention and group therapy (Weiss, 1994). |
| |
The LSES has been used in numerous unpublished master's
theses as well as a study conducted by Jacobson (1993) (cited in: M.J.
Salamon, personal communication, April, 1999). In Jacobson's study,
Salamon noted that the LSES was used along with the State-Trait Anxiety
Inventory with women of the baby-boomer age group to assess their level
of depression and their perceived satisfaction in life (M.J. Salamon,
personal communication, April, 1999). These various applications of the
LSES instrument in research strengthen the validity of the LSES as an
adequate research and assessment tool of individuals across the ages. |
| Good interpretation |
Test Bias
Closer scrutiny of the normative sample group, by Busch-Rossnagel,
raised questions as to the omission of data regarding gender breakdown,
education level, and socioeconomic status (Conoley et al., 1988). An
additional criticism mentioned by various authors is that the instrument
appears to be normed on a small cross-section of a portion of the North
Eastern US, which would indicate that this normative sample does not
reflect the greater population (Conoley et al., 1988; Neri, Andermarcher,
Pradelli & Salvioli, 1995; Phillips, 1986). |
|
Add year after 'Dixon and Busch-Rossnagle'
|
Discussion
The LSES was developed by Salamon and Conte (1984) in order to
provide the social and gerontological sciences with an improved
instrument used to measure the abstract concept of life satisfaction.
Dixon and Busch-Rossnagle state that the authors have adequately
designed this instrument with internal consistency levels measuring
stronger psychometric properties in the moderate to high range depending
on the subscale. They also expanded the pre-existing life satisfaction
measurement scales to include additional characteristics of concepts of
satisfaction in life that the elderly individuals reported to be
important at this stage in life.(Conoley et al., 1988). The LSES
instrument appears easy to use, inexpensive, and is considered non
threatening by the elderly, advised Salamon (1987). |
| |
Both Dixon and Busch-Rossnagel caution that interpretation
must be made with discretion since the instrument is tapping into one's
subjective self-report assessment of mood and attitude in the moment (Conoley
et al., 1988). |
| |
Salamon (1988) reported two cases in which he used the
LSES in the clinical setting; in each case, the tool was used for
assessment prior to treatment for depression and post-treatment. In both
instances, he noted marked improvement in subscale scores and recommends
that this instrument could best be used as an assessment measure (p.
51-53). He and others have also argued that this instrument has shown
relevance as a research tool and also has been used successfully as a
springboard for dialog during the clinical intake procedure, focusing on
areas of specific topics (i.e., low scores in subscales of Goodness of
fit , Mood tone, Meaningfulness of life and etc.) and for strengthening
rapport with the elderly client (e.g., Dean-Church & Gilroy, 1993;
Gould, 1992; Neri et al., 1995; Phillips, 1986; Weiss, 1994) |
| |
Realizing that the construct of life satisfaction is
abstract and difficult to measure due to variables such as the
individual's frame of reference, culture, ethnicity, values and
spirituality, the LSES should never be used as an absolute measure of
life satisfaction. It does, however, show remarkable value as a tool
used in the assessment of depression and as a pre and post treatment
assessment tool, however it is the opinion of this reviewer that the
LSES should be administered to a larger, stratified sample group in
order to further improve the psychometric properties. |
|
Add a space between all the authors' initials,
for example, 'Barrow, G. M.'
Underline 'Educational'
|
References
Barrow, G.M., (1996). Special problems. Aging, the Individual and
Society (6th ed.). Minneapolis/St. Paul: West Publishing Co.
Brockett, R.G., (1987). Life satisfaction and learner self-direction:
enhancing quality of life during the later years. Educational
Gerontology, 13, 225-237.
Cohen, R.J., Swerdilk, M.E., Phillips, S.M., (Eds.). (1996). Psychological
Testing and Assessment: An introduction to tests and measurements (3rd
ed.). Mountain View, CA: Mayfield Publishing Company.
Conoley, J.C., Kramer, J.J., & Mitchell, J.V. (Eds.). (1988). The
Supplement to the Ninth Mental Measurements Yearbook. Lincoln, NE:
University of Nebraska Press. (pp.170-172).
Conte, V.A., & Salamon, M.J., (1982). An objective approach to
the measurement and use of life satisfaction with older persons. Measurement
and Evaluation in Guidance, 15, (3): 194-200.
Dean-Church, L., & Girlie, F.D., (1993). Relation of sex-role
orientation to life satisfaction in a healthy elderly sample. Journal
of Social Behavior and Personality, 8, (1) 133-140.
Euler, B.L., (1992). A flaw in gerontological assessment: the weak
relationship of elderly superficial life satisfaction to deep
psychological well-being. The International Journal of Aging and
Human Development, 34, (4) 299-310.
Gould, M.T., (1992). Nursing home elderly: social-environmental
factors. Journal of Gerontological Nursing,18, (8)13-20.
Neri, M., Andermarcher, E., Pradelli, J.M., & Salvioli, G.,
(1995). Influence of a double blind pharmacological trial on two domains
of well-being in subjects with age associated memory impairment. Archives
of Gerontology and Geriatrics, 21, (3) 241-252.
Neugarten, B.L., Havighurst, R.J., & Tobin, S. S., (1961). The
measurement of life satisfaction. The Journal of Gerontology, 16,
134-143.
Phillips, E.J., (1986). A review of the life satisfaction in the
elderly scale. Journal of Counseling and Development, 64:
542-543.
Rathus, S.A., (1993). Psychology (5th ed.). Philadelphia, PA:
Harcourt Brace College Publishers.
Ryff, C.D., (1995). Psychological well-being in adult life. Current
Directions in Psychological Science, 4, (4) 99-104.
Salamon, M.J., (1985). Sociocultural role theories in the elderly: a
replication and extension. Activities, Adaptation and Aging, 7,
(2) 111-122.
Salamon, M.J., (1987). Health care environment and life satisfaction
in the elderly. Journal of Aging Studies, 1, (3) 287-297.
Salamon, M.J., (1988). Clinical use of the life satisfaction in the
elderly scale. Clinical Gerontologist, 8, (1) 45-54.
Schultz, R., Heckhausen, J., (1996). A life span model of successful
aging. American Psychologist, 702-714.
Weiss, J.C., (1994). Group therapy with older adults in long-term
care settings: research and clinical cautions and recommendations. The
Journal for Specialists in Group Work, 19, (1) 22-29 |
|
