Journal of Gerontology 1976. Vol. 31, No. I. 89-98
Psychological and Social Correlates of Life Satisfaction as a Function of Residential Constraint Stephen Wolk, PhD,1 and Sharon Telleen, MED1
SYCHOLOGICAL investigations of P aging have considered life satisfaction to be an important component of successful aging (Tobin & Neugarten, 1961). In an attempt to understand this aspect of aging, research has examined various correlates of life satisfaction. These have usually included health, activity level, formal and informal group participation, and education and income levels (Edwards & Klemmack, 1973). Of all the correlates studied, self-assessed health has been found to be one of the most significant predictors of life satisfaction (Palmore & Luikart, 1972; Edwards & Klemmack, 1973; Spreitzer, 1974). Generally, the healthier one perceives himself the greater the life satisfaction reported. Similarly, the level of activity maintained both formally and informally has also been found to be an important correlate of satisfaction (Bley, Goodman, Dye, & Harel, 1972; Lemon, Bengtson, & Peterson, 1972; Tobin & Neugarten, 1961). However, it is noteworthy that, when health and socioeconomic levels are controlled, organized voluntary participation no longer correlates with satisfaction (Bull & Aucoin, 1975; Edwards & Klemmack, 1973). Apparently, perceived health is the more basic component to satisfaction, i.e., if one feels
•Institute for Child Study, Univ. of Maryland, College Park 20742. The authors wish to express their appreciation to Dr. John Kurtz for his assistance and guidance.
healthy, then one is likely to be more active, at least in regard to formal associations, which eventuates in satisfaction with life. Somewhat analogously, economic sufficiency, at least for some aged populations, also relates in a basic fashion to life satisfaction (Spreitzer, 1974). More recently the relationship between life satisfaction and developmental task accomplishment has been investigated (Kurtz & Wolk, 1975). The concept of the developmental task defines a particular age-related challenge, biological, psychological, or social in nature, which requires a degree of successful resolution for development to proceed. Associated with the late years is a set of subtasks (for example, learning to get along without one's deceased spouse) (Havighurst, 1972). It was found that life satisfaction was greater for those elderly persons who reported more success in dealing with such tasks. Although several social-psychological variables seem to be documented as important correlates of successful aging, it remains to be evidenced whether these variables consistently relate to satisfaction across the types of settings in which elderly often reside. Equally so the research considering life satisfaction should begin to evaluate whether various settings also produce in their residents varying levels of satisfaction. There are two points being raised: Does the nature of the setting have some influence upon the over-all level of life 89
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
Two questions were addressed in the present study: Do residential settings of varying levels of constraint influence life satisfaction? Do such settings involve different correlates of life satisfaction? A total of 129 male and female ambulatory residents were surveyed in settings of high and low constraint on the following measures: life satisfaction, developmental task resolution, self-acceptance, perceived autonomy, activity level, health and educational level. Stepwise regression, covariance, and t-test analyses indicated: (1) life satisfaction and developmental task accomplishment were greater in the lower constraining setting; (2) the selected correlates resulted in multiple correlations of .675 and .590 with satisfaction; (3) differing sets of correlates significantly predicted satisfaction in each setting: health the most important in the high constraining setting and perceived autonomy and selfconcept important to the low constraining setting; (4) developmental task success significantly predicted satisfaction in both settings.
90
WOLKAND TELLEEN their apparent discrepancy in degree of environmental constraint. Consistent with the findings of Smith and Lipman (1972) and the argument of Lawton and Cohen (1974) it was predicted that the average level of life satisfaction would be greater in the lower constraining setting. Additionally the following question was posed: Do the variables of health, developmental task accomplishment, self-concept, activity level, and perceived autonomy differentially predict life satisfaction in settings of varying levels of environmental constraint? METHOD
Sample
Setting A. — Represented in the setting was a sectarian sponsored (Lutheran) "retirement home." Located in a suburban area, this setting contains a group of elderly who, while possessing separate rooms in the living quarters, have all basic needs met through organized staff procedures. Rules are well defined and a day's schedule is rigidly followed. Subjects consisted of 51 individuals (16 males, 35 females) with a median age of 77. All were Caucasian. The sample represented approximately 35% of the total number of ambulatory residents. Setting B. — This residence setting represented a retirement-type village in which residents purchase their own homes.2 The elderly lead an independent existence; all basic needs are met individually, and individuals live in separate dwellings. Many activities are sponsored by resident clubs and groups, although no specific rules govern participation in any of these activities. The village is located in a suburban area. Subjects from this setting consisted of 78 individuals (31 males, 47 females). All subjects were Caucasian. The median age was 74 years. The sample represented 25% of the total residential community. Since the sample from each setting did not represent a truly random selection from the potential populations of elderly, some mention should be made of the degree to which sampling bias was kept at a minimum. It is contended that such bias was reduced through the manner in which subjects were solicited
"The village is a local version of Rossmoor Leisure World.
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
satisfaction? Do various types of settings involve different determinants (the kinds of social-psychological factors noted above) of life satisfaction? Indeed, it is the case that a great proportion of research cited previously has failed to define systematically the setting from which populations were identified and evaluated such settings across assessments. What parameters of the environmental setting might influence the level and correlates of life satisfaction? This question has been pursued in a general way from a variety of directions. Lawton, Nahemow, & Teaff (1975) found that tenants in housing located in small communities had higher activity participation, greater contact with friends, and greater housing satisfaction than tenants in larger communities. Those elderly residing in highrise buildings manifested lower housing satisfaction and lower mobility, which suggests that some restriction of life space may be greater in high rise buildings. Maxwell, Bader, and Watson (1972) have argued that institutional type settings may contain numerous sources of constraint, including physical restraint, psychopharmacological restraint (drugs, etc.) and social restraint. Indeed, Lawton and Cohen (1974) have maintained that the environment, with its variations in constraint, has the potential for influencing the life satisfaction of elderly, who, more so than general populations, may be influenced by environmental factors. Smith and Lipman (1972) have examined environmental constraint, which they defined as an individual's ability (1) to perform various self-care tasks, (2) to move about, and (3) to be gainfully employed. These investigators concluded from their data that residents who were described as constrained were less likely to be satisfied than residents who were not constrained. The present study considers environmental constraint an important parameter of most elderly settings and worthy of additional consideration. Specifically the perception of psychological and social restrictions, beyond basic economic constraint, would also seem important to the development and maintenance of such psychological states as satisfaction and adjustment. In conclusion, the purposes of this investigation are to assess the level of life satisfaction and to ascertain the most important correlates of satisfaction in two environments. The settings were chosen for
LIFE SA TISFA CTIONAND RESIDENTIAL CONSTRAINT
must be supplied by the resident. It is not the case that Setting A residents are financially prohibited from residing in a more autonomous situation such as Setting B, since costs are highly comparable. It was indicated by several of the professional staff of Setting A that indigent elderly reside at publicly supported retirement homes in the metropolitan area, and that residents must satisfy all economic obligations prior to acceptance for residency. Instrumentation Life satisfaction. — The Life Satisfaction Index A, an 18-item questionnaire, (Neugarten, Havighurst, & Tobin, 1961), as modified by Adams (1969), was used to assess the subject's characterization of his emotional and social adjustment to the late years. Each item asks for agreement or disagreement and a total score (0-18) is derived which represents degree of satisfaction. The scale has been used quite extensively in gerontological research (Adams, 1969). Examples of several of the items are: "This is the dreariest time of my life"; "I feel old and somewhat tired"; and "As I look back on my life, I am fairly well satisfied." Wolk & Kurtz (1975) report a reliability coefficient of .84 (internal consistency). Developmental task adjustment. — An application of the developmental conceptualization of aging is Havighurst's (1972) developmental task. Havighurst defines living as a long series of tasks where learning them well brings satisfaction and reward but learning them poorly brings unhappiness and social disapproval. Such tasks revolve around issues that constitute healthy and satisfactory growth, analogous to the concept of critical periods of organ formation in embryonic development. Developmental tasks arise from a combination, of interpersonal processes and cultural factors. Specifically with respect to later maturity, the developmental tasks of this age reflect what Havighurst termed a "defensive strategy." The aged individual grows limited in the physical, mental, and economic spheres. But, he will likely experience a number of crises such as decreased income, loss of a spouse by death, or a crippling illness, and he must work harder to hold on to what he already has rather than try to acquire more.
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
for participation. In Setting A the total set of instruments was dispensed during a time (dinner) when all ambulatory residents gathered in one location. Each resident received a set and had the option to participate. There is no reason to believe that those who did not opt to take part were less healthy or active than those elderly who did respond, given the nature of the demands for participation and the conditions under which the instruments were distributed. Activity and group involvement in Setting B revolve around organized club functions. It was necessary to contact individuals in such a setting. Therefore a general club function was selected which did not concern a political, athletic or sex-related activity, so that a more representative group of elderly would be in attendance. With the assistance of club leaders all members of this club (American Association of Retired Persons) including those not in attendance were given or sent copies of the instruments. Again participation was invited, and there is no indication that those who did not volunteer represented a segment of the group that was either less healthy or active. No direct measure of socioeconomic status was obtained for the residents of each setting. However, several indirect indices would suggest that the residents of each setting were comparable in socioeconomic background. First, an assessment was made of the general breakdown in levels of employment prior to retirement. (It should be noted that over 95% of the residents in each setting were retired.) Approximately 3 out of every 5 residents in Setting A and 2 out of every 3 in Setting B had been employed at occupations that could be labeled professional or white collar. Second, information was gathered in regard to the economic demands of living in each setting. The residents of Setting B purchase their own homes and are responsible for providing their own food, clothing, medical treatment, and other commodities. The approximate average cost of financing a dwelling unit and providing for basic needs is $800.00 per month. It should be noted that for those individuals still married, the cost per individual is somewhat less. Setting A represents a moderately expensive retirement home. The elderly are responsible for meeting an average of $700 to $750 per month fee. This fee includes room, board, club activity expenses, and medical treatment. All other commodities
91
92
WOLKAND TELLEEN taken as one of the important correlates of life satisfaction. An attempt was made, therefore, to phrase items that conveyed a description of behavior or a disposition to behave that were literal to each developmental task. Admittedly, several of the items (e.g., #6, 8, 10, 14) include some affective content that could be confounded with a measure of satisfaction. Thus, any empirical relationship obtained between these measures would require at least some qualification. Hopefully, future research and analysis will generate a more independent measure of developmental task accomplishment. A total score (14-70) was derived to reflect degree of successful developmental task accomplishment. Self-concept. — The Self-Acceptance Scale (Berger, 1952) assessed self-concept. The concept can be operationally defined as an assessment of "a faith in capacity to cope with life, responsibility, objective acceptance of criticism, sense of self-worth, and an absence of shyness or self-consciousness" (Robinson & Shaver, 1972). The instrument contains 36 items, requiring responses on a five-point Likert scale ("not at all true of myself" to "true of myself"). Whole-test reliability coefficients are reported to be .89 and .75 (Robinson & Shaver, 1972) for the scale. A total score (36-180) is derived, and is scored in the direction of low self-acceptance (high score). Activity level. — The measurement of the day-to-day involvement level maintained by the subject necessitated the development of a check list of activities potentially available in each setting. A separate list was developed for each setting and contained 40 activities of both a formal and informal nature. For each activity, a subject checked whether his involvement was "regular" (2 points), "irregular" (1 point), or "none" (0 points). A total score on the index (0-80) thus indicated the breadth and level of activity. Environmental constraint. — Included in the battery was an assessment of a subject's perception of the environment in regard to the presence of constraining rules, the level of personal autonomy, and the role of the individual in determining policy. Although the settings themselves were selected on the basis of the apparent differences in level of constraint, this measure was deemed necessary for two reasons: (1) it served as a partial check of the assumed differences between set-
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
In one task the individual is faced with adjusting to decreasing physical strength and health. This suggests that the individual no longer has the stamina and capacity to do very physical and strenuous activities as he had in earlier years. And, as one adjusts to the increased susceptibility to illness, one must often adjust to invalidism as a result of disease or accident. With the advent of retirement and a subsequent reduction in income, one's activities and life style must frequently change. The task then is to adjust to retirement and reduced income by finding new ways of occupying one's time and adjusting one's outlook by redefining goals and orientations. Another task centers upon adjusting to the death of one's spouse. A life style change resulting in loneliness typically occurs, which necessitates the learning of new ways to fulfill one's needs. One must come to admit finally that he is old. He must come to establish an explicit affiliation with his age group. It becomes necessary to accept one's status as an elderly citizen and function constructively in the "old age" group. The individual must learn to meet social and civic obligations. This may typically occur through the avenues of adult education programs and civic and political groups. One must also establish satisfactory living arrangements. The type of housing that one chooses should be responsive to the special physical, psychological, and social needs of the elderly. Based upon these developmental tasks of the late years, an index of developmental adjustment was established. This index assessed the degree of success in meeting these developmental tasks and included 2 items for each of the 7 developmental tasks, for a total of 14 items (appendix). The items covered both attitudinal and behavioral dimensions of the tasks, and the subject was required to indicate on a five-point Likert scale his degree of task accomplishment ("not at all true of myself" to "completely true of myself"). This index represented a minor revision of one used by Wolk and Kurtz (1975), who report an internal consistency coefficient of .61 for the index. A note is in order concerning the possible overlap in item content between the measure of life satisfaction and that for developmental task adjustment. Conceptually, developmental task accomplishment is
LIFE SA TISFACTIONAND RESIDENTIAL CONSTRAINT
ducted to analyze potential setting differences in regard to life satisfaction. In order to investigate the relationship between selected correlates and life satisfaction for each setting two analyses were conducted: calculation of zero order Pearson productmoment coefficients of correlation between variables; computation of a sequence of multiple linear regression equations in a stepwise manner. The latter analysis was selected to delineate more fully the strength and relative independence of the various correlates for the prediction of life satisfaction in each setting. RESULTS
Table 1 presents the means, standard deviations, and differences between residential settings on all assessment variables. In regard to the prediction that greater life satisfaction would be manifested by residents in the lower constraining setting (Setting B) a significant difference (p < .01) in life satisfaction scores in favor of Setting B was obtained. However, it should also be noted (Table 1) that the elderly differed significantly on four other variables between residences: education and health levels, developmental task accomplishment, and perceived environmental autonomy. The latter of these differences would seem to support in part the assumed levels of constraint present in each residence. However, it is apparent that the settings could have differed much more on this rating than was observed. Two possible explanations can be offered. It may be that the type of items included in the measure of environmental constraint represent only a segment of the differences between settings, although the content of the items was established with the aid of residents and staff of both settings. Procedure The various assessments were collated into Somewhat alternatively, the settings may not a self-contained booklet. In both settings inTable 1. Means and Standard Deviations for all Variables dividuals responded during a general meeting by Setting and t Values of Differences Between Means. session. Responses were given anonymously. No individual was surveyed who was inResidence A t Value Residence B capacitated and/or who currently required S.D. S.D. X X custodial care. Variable (N =:78) ( N - = 51) 8.87* 3.82 1.11 1.73 5.95 For each setting means and standard devia- Educational level .67 5.16* health .78 3.53 4.15 tions were calculated for all assessed Perceived Environmental autonomy 3.69* 3.59 3.43 22.45 24.78 variables. To ascertain over-all setting dif- Self-acceptance 20.39 16.75 1.75 70.94 64.94 6.12* 3.59 2.53 11.00 14.49 ferences, independent t-tests of differences Life satisfaction Developmental task between means were computed. In addition, a accomplishment 4.97* 7.56 7.65 50.43 57.19 9.14 one-way analysis of covariance, using health Activity level 8.42 .01 20.25 20.23 and education ratings as covariates and life satisfaction as the criterion measure, was con- ' /K.01, two-tailed test.
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
tings, for the individuals surveyed from each setting and (2) it afforded a measure of perceived constraint that could be related within each setting to life satisfaction. Each of the six items required a response on a five-point Likert scale from "not at all true" to "completely true." The index was scored in the direction of perceived low situational constraint. The items were: "The rules here keep one from doing the things one wants to do"; "living here makes one dependent upon others"; "Residents are asked for their advice when changes are planned and carried out by administrators"; "Residents have free access to facilities and grounds"; "The staff here responds to residents' requests"; and "If a resident has a problem, he can solve it by taking the initiative." A total score (5-30) was derived. Perceived health and educational level. —For self-perceived health the measure was very similar to that employed by Palmore and Luikart (1972) and Edwards and Klemmack (1973). It consisted of a direct question in regard to perceived health with five alternatives: "Would you say your own health in general is: excellent; good; fair; poor; very poor?" The response to the question was coded from 5 (excellent) to 1 (very poor). The measure of educational level consisted of 7 choices: "My formal education consists of: 1-6 years (elementary school); 7-9 years Gunior high school); 10-11 years (1 or 2 years of high school); 12 years (high school diploma); 13-15 years (several years of college); 16 years (college diploma); 16+ years (graduate school). The response was coded from 1 (1-6 years) to 7 (16+ years).
93
94
WOLKAND TELLEEN The obtained difference between settings for developmental task accomplishment will be discussed in partial support of the role of environmental constraint for over-all adjustment and satisfaction. Indeed, as reported in Table 2, developmental task accomplishment correlated significantly (p < .01) with the degree of perceived autonomy in Setting B. In Setting A perceived health, however, not autonomy, related significantly (p < .05) to developmental task accomplishment. Finally, it is somewhat noteworthy that, although the settings were selected for and did manifest differing levels of psychological and social constraint, the residents of each setting did not report discrepant levels of participation in both formal and informal activities. The data presented in Tables 2 and 3 relate to the investigation of similarities and differences in the correlates of life satisfaction as a function of residential constraint. Table 2 presented the intercorrelation matrix of all variables for each setting. The potential significance of these coefficients resides in an examination of the significant correlates of life satisfaction for each setting. For Setting A these were: perceived health, perceived environmental autonomy, developmental task
Table 2. Correlation Matrix of All Variables by Setting. 3
2
1
Variable
-.24
(1) Education .01
(2) Health
4
5
6
7
-.25
-.20
.21
-.01
-.24
.21
-.02
.48*
.30"
.14
-.08
.31"
.12
.02
-.17
.06
(3) Environmental autonomy
-.10
.08
(4) Self-acceptance
-.18
.12
-.25"
(S) Life satisfaction
.07
.10
.42*
-.38*
(6) Developmental Task
.02
.10
.34*
-.37*
.45*
(7) Activity level
.02
-.12
.07
.04
.16
-.14
.43*
.33" .20
.15
Note: Matrix above the diagonal represents Residence A, that below the diagonal Residence B. V
Psychological and Social Correlates of Life Satisfaction as a Function of Residential Constraint Stephen Wolk, PhD,1 and Sharon Telleen, MED1
SYCHOLOGICAL investigations of P aging have considered life satisfaction to be an important component of successful aging (Tobin & Neugarten, 1961). In an attempt to understand this aspect of aging, research has examined various correlates of life satisfaction. These have usually included health, activity level, formal and informal group participation, and education and income levels (Edwards & Klemmack, 1973). Of all the correlates studied, self-assessed health has been found to be one of the most significant predictors of life satisfaction (Palmore & Luikart, 1972; Edwards & Klemmack, 1973; Spreitzer, 1974). Generally, the healthier one perceives himself the greater the life satisfaction reported. Similarly, the level of activity maintained both formally and informally has also been found to be an important correlate of satisfaction (Bley, Goodman, Dye, & Harel, 1972; Lemon, Bengtson, & Peterson, 1972; Tobin & Neugarten, 1961). However, it is noteworthy that, when health and socioeconomic levels are controlled, organized voluntary participation no longer correlates with satisfaction (Bull & Aucoin, 1975; Edwards & Klemmack, 1973). Apparently, perceived health is the more basic component to satisfaction, i.e., if one feels
•Institute for Child Study, Univ. of Maryland, College Park 20742. The authors wish to express their appreciation to Dr. John Kurtz for his assistance and guidance.
healthy, then one is likely to be more active, at least in regard to formal associations, which eventuates in satisfaction with life. Somewhat analogously, economic sufficiency, at least for some aged populations, also relates in a basic fashion to life satisfaction (Spreitzer, 1974). More recently the relationship between life satisfaction and developmental task accomplishment has been investigated (Kurtz & Wolk, 1975). The concept of the developmental task defines a particular age-related challenge, biological, psychological, or social in nature, which requires a degree of successful resolution for development to proceed. Associated with the late years is a set of subtasks (for example, learning to get along without one's deceased spouse) (Havighurst, 1972). It was found that life satisfaction was greater for those elderly persons who reported more success in dealing with such tasks. Although several social-psychological variables seem to be documented as important correlates of successful aging, it remains to be evidenced whether these variables consistently relate to satisfaction across the types of settings in which elderly often reside. Equally so the research considering life satisfaction should begin to evaluate whether various settings also produce in their residents varying levels of satisfaction. There are two points being raised: Does the nature of the setting have some influence upon the over-all level of life 89
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
Two questions were addressed in the present study: Do residential settings of varying levels of constraint influence life satisfaction? Do such settings involve different correlates of life satisfaction? A total of 129 male and female ambulatory residents were surveyed in settings of high and low constraint on the following measures: life satisfaction, developmental task resolution, self-acceptance, perceived autonomy, activity level, health and educational level. Stepwise regression, covariance, and t-test analyses indicated: (1) life satisfaction and developmental task accomplishment were greater in the lower constraining setting; (2) the selected correlates resulted in multiple correlations of .675 and .590 with satisfaction; (3) differing sets of correlates significantly predicted satisfaction in each setting: health the most important in the high constraining setting and perceived autonomy and selfconcept important to the low constraining setting; (4) developmental task success significantly predicted satisfaction in both settings.
90
WOLKAND TELLEEN their apparent discrepancy in degree of environmental constraint. Consistent with the findings of Smith and Lipman (1972) and the argument of Lawton and Cohen (1974) it was predicted that the average level of life satisfaction would be greater in the lower constraining setting. Additionally the following question was posed: Do the variables of health, developmental task accomplishment, self-concept, activity level, and perceived autonomy differentially predict life satisfaction in settings of varying levels of environmental constraint? METHOD
Sample
Setting A. — Represented in the setting was a sectarian sponsored (Lutheran) "retirement home." Located in a suburban area, this setting contains a group of elderly who, while possessing separate rooms in the living quarters, have all basic needs met through organized staff procedures. Rules are well defined and a day's schedule is rigidly followed. Subjects consisted of 51 individuals (16 males, 35 females) with a median age of 77. All were Caucasian. The sample represented approximately 35% of the total number of ambulatory residents. Setting B. — This residence setting represented a retirement-type village in which residents purchase their own homes.2 The elderly lead an independent existence; all basic needs are met individually, and individuals live in separate dwellings. Many activities are sponsored by resident clubs and groups, although no specific rules govern participation in any of these activities. The village is located in a suburban area. Subjects from this setting consisted of 78 individuals (31 males, 47 females). All subjects were Caucasian. The median age was 74 years. The sample represented 25% of the total residential community. Since the sample from each setting did not represent a truly random selection from the potential populations of elderly, some mention should be made of the degree to which sampling bias was kept at a minimum. It is contended that such bias was reduced through the manner in which subjects were solicited
"The village is a local version of Rossmoor Leisure World.
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
satisfaction? Do various types of settings involve different determinants (the kinds of social-psychological factors noted above) of life satisfaction? Indeed, it is the case that a great proportion of research cited previously has failed to define systematically the setting from which populations were identified and evaluated such settings across assessments. What parameters of the environmental setting might influence the level and correlates of life satisfaction? This question has been pursued in a general way from a variety of directions. Lawton, Nahemow, & Teaff (1975) found that tenants in housing located in small communities had higher activity participation, greater contact with friends, and greater housing satisfaction than tenants in larger communities. Those elderly residing in highrise buildings manifested lower housing satisfaction and lower mobility, which suggests that some restriction of life space may be greater in high rise buildings. Maxwell, Bader, and Watson (1972) have argued that institutional type settings may contain numerous sources of constraint, including physical restraint, psychopharmacological restraint (drugs, etc.) and social restraint. Indeed, Lawton and Cohen (1974) have maintained that the environment, with its variations in constraint, has the potential for influencing the life satisfaction of elderly, who, more so than general populations, may be influenced by environmental factors. Smith and Lipman (1972) have examined environmental constraint, which they defined as an individual's ability (1) to perform various self-care tasks, (2) to move about, and (3) to be gainfully employed. These investigators concluded from their data that residents who were described as constrained were less likely to be satisfied than residents who were not constrained. The present study considers environmental constraint an important parameter of most elderly settings and worthy of additional consideration. Specifically the perception of psychological and social restrictions, beyond basic economic constraint, would also seem important to the development and maintenance of such psychological states as satisfaction and adjustment. In conclusion, the purposes of this investigation are to assess the level of life satisfaction and to ascertain the most important correlates of satisfaction in two environments. The settings were chosen for
LIFE SA TISFA CTIONAND RESIDENTIAL CONSTRAINT
must be supplied by the resident. It is not the case that Setting A residents are financially prohibited from residing in a more autonomous situation such as Setting B, since costs are highly comparable. It was indicated by several of the professional staff of Setting A that indigent elderly reside at publicly supported retirement homes in the metropolitan area, and that residents must satisfy all economic obligations prior to acceptance for residency. Instrumentation Life satisfaction. — The Life Satisfaction Index A, an 18-item questionnaire, (Neugarten, Havighurst, & Tobin, 1961), as modified by Adams (1969), was used to assess the subject's characterization of his emotional and social adjustment to the late years. Each item asks for agreement or disagreement and a total score (0-18) is derived which represents degree of satisfaction. The scale has been used quite extensively in gerontological research (Adams, 1969). Examples of several of the items are: "This is the dreariest time of my life"; "I feel old and somewhat tired"; and "As I look back on my life, I am fairly well satisfied." Wolk & Kurtz (1975) report a reliability coefficient of .84 (internal consistency). Developmental task adjustment. — An application of the developmental conceptualization of aging is Havighurst's (1972) developmental task. Havighurst defines living as a long series of tasks where learning them well brings satisfaction and reward but learning them poorly brings unhappiness and social disapproval. Such tasks revolve around issues that constitute healthy and satisfactory growth, analogous to the concept of critical periods of organ formation in embryonic development. Developmental tasks arise from a combination, of interpersonal processes and cultural factors. Specifically with respect to later maturity, the developmental tasks of this age reflect what Havighurst termed a "defensive strategy." The aged individual grows limited in the physical, mental, and economic spheres. But, he will likely experience a number of crises such as decreased income, loss of a spouse by death, or a crippling illness, and he must work harder to hold on to what he already has rather than try to acquire more.
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
for participation. In Setting A the total set of instruments was dispensed during a time (dinner) when all ambulatory residents gathered in one location. Each resident received a set and had the option to participate. There is no reason to believe that those who did not opt to take part were less healthy or active than those elderly who did respond, given the nature of the demands for participation and the conditions under which the instruments were distributed. Activity and group involvement in Setting B revolve around organized club functions. It was necessary to contact individuals in such a setting. Therefore a general club function was selected which did not concern a political, athletic or sex-related activity, so that a more representative group of elderly would be in attendance. With the assistance of club leaders all members of this club (American Association of Retired Persons) including those not in attendance were given or sent copies of the instruments. Again participation was invited, and there is no indication that those who did not volunteer represented a segment of the group that was either less healthy or active. No direct measure of socioeconomic status was obtained for the residents of each setting. However, several indirect indices would suggest that the residents of each setting were comparable in socioeconomic background. First, an assessment was made of the general breakdown in levels of employment prior to retirement. (It should be noted that over 95% of the residents in each setting were retired.) Approximately 3 out of every 5 residents in Setting A and 2 out of every 3 in Setting B had been employed at occupations that could be labeled professional or white collar. Second, information was gathered in regard to the economic demands of living in each setting. The residents of Setting B purchase their own homes and are responsible for providing their own food, clothing, medical treatment, and other commodities. The approximate average cost of financing a dwelling unit and providing for basic needs is $800.00 per month. It should be noted that for those individuals still married, the cost per individual is somewhat less. Setting A represents a moderately expensive retirement home. The elderly are responsible for meeting an average of $700 to $750 per month fee. This fee includes room, board, club activity expenses, and medical treatment. All other commodities
91
92
WOLKAND TELLEEN taken as one of the important correlates of life satisfaction. An attempt was made, therefore, to phrase items that conveyed a description of behavior or a disposition to behave that were literal to each developmental task. Admittedly, several of the items (e.g., #6, 8, 10, 14) include some affective content that could be confounded with a measure of satisfaction. Thus, any empirical relationship obtained between these measures would require at least some qualification. Hopefully, future research and analysis will generate a more independent measure of developmental task accomplishment. A total score (14-70) was derived to reflect degree of successful developmental task accomplishment. Self-concept. — The Self-Acceptance Scale (Berger, 1952) assessed self-concept. The concept can be operationally defined as an assessment of "a faith in capacity to cope with life, responsibility, objective acceptance of criticism, sense of self-worth, and an absence of shyness or self-consciousness" (Robinson & Shaver, 1972). The instrument contains 36 items, requiring responses on a five-point Likert scale ("not at all true of myself" to "true of myself"). Whole-test reliability coefficients are reported to be .89 and .75 (Robinson & Shaver, 1972) for the scale. A total score (36-180) is derived, and is scored in the direction of low self-acceptance (high score). Activity level. — The measurement of the day-to-day involvement level maintained by the subject necessitated the development of a check list of activities potentially available in each setting. A separate list was developed for each setting and contained 40 activities of both a formal and informal nature. For each activity, a subject checked whether his involvement was "regular" (2 points), "irregular" (1 point), or "none" (0 points). A total score on the index (0-80) thus indicated the breadth and level of activity. Environmental constraint. — Included in the battery was an assessment of a subject's perception of the environment in regard to the presence of constraining rules, the level of personal autonomy, and the role of the individual in determining policy. Although the settings themselves were selected on the basis of the apparent differences in level of constraint, this measure was deemed necessary for two reasons: (1) it served as a partial check of the assumed differences between set-
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
In one task the individual is faced with adjusting to decreasing physical strength and health. This suggests that the individual no longer has the stamina and capacity to do very physical and strenuous activities as he had in earlier years. And, as one adjusts to the increased susceptibility to illness, one must often adjust to invalidism as a result of disease or accident. With the advent of retirement and a subsequent reduction in income, one's activities and life style must frequently change. The task then is to adjust to retirement and reduced income by finding new ways of occupying one's time and adjusting one's outlook by redefining goals and orientations. Another task centers upon adjusting to the death of one's spouse. A life style change resulting in loneliness typically occurs, which necessitates the learning of new ways to fulfill one's needs. One must come to admit finally that he is old. He must come to establish an explicit affiliation with his age group. It becomes necessary to accept one's status as an elderly citizen and function constructively in the "old age" group. The individual must learn to meet social and civic obligations. This may typically occur through the avenues of adult education programs and civic and political groups. One must also establish satisfactory living arrangements. The type of housing that one chooses should be responsive to the special physical, psychological, and social needs of the elderly. Based upon these developmental tasks of the late years, an index of developmental adjustment was established. This index assessed the degree of success in meeting these developmental tasks and included 2 items for each of the 7 developmental tasks, for a total of 14 items (appendix). The items covered both attitudinal and behavioral dimensions of the tasks, and the subject was required to indicate on a five-point Likert scale his degree of task accomplishment ("not at all true of myself" to "completely true of myself"). This index represented a minor revision of one used by Wolk and Kurtz (1975), who report an internal consistency coefficient of .61 for the index. A note is in order concerning the possible overlap in item content between the measure of life satisfaction and that for developmental task adjustment. Conceptually, developmental task accomplishment is
LIFE SA TISFACTIONAND RESIDENTIAL CONSTRAINT
ducted to analyze potential setting differences in regard to life satisfaction. In order to investigate the relationship between selected correlates and life satisfaction for each setting two analyses were conducted: calculation of zero order Pearson productmoment coefficients of correlation between variables; computation of a sequence of multiple linear regression equations in a stepwise manner. The latter analysis was selected to delineate more fully the strength and relative independence of the various correlates for the prediction of life satisfaction in each setting. RESULTS
Table 1 presents the means, standard deviations, and differences between residential settings on all assessment variables. In regard to the prediction that greater life satisfaction would be manifested by residents in the lower constraining setting (Setting B) a significant difference (p < .01) in life satisfaction scores in favor of Setting B was obtained. However, it should also be noted (Table 1) that the elderly differed significantly on four other variables between residences: education and health levels, developmental task accomplishment, and perceived environmental autonomy. The latter of these differences would seem to support in part the assumed levels of constraint present in each residence. However, it is apparent that the settings could have differed much more on this rating than was observed. Two possible explanations can be offered. It may be that the type of items included in the measure of environmental constraint represent only a segment of the differences between settings, although the content of the items was established with the aid of residents and staff of both settings. Procedure The various assessments were collated into Somewhat alternatively, the settings may not a self-contained booklet. In both settings inTable 1. Means and Standard Deviations for all Variables dividuals responded during a general meeting by Setting and t Values of Differences Between Means. session. Responses were given anonymously. No individual was surveyed who was inResidence A t Value Residence B capacitated and/or who currently required S.D. S.D. X X custodial care. Variable (N =:78) ( N - = 51) 8.87* 3.82 1.11 1.73 5.95 For each setting means and standard devia- Educational level .67 5.16* health .78 3.53 4.15 tions were calculated for all assessed Perceived Environmental autonomy 3.69* 3.59 3.43 22.45 24.78 variables. To ascertain over-all setting dif- Self-acceptance 20.39 16.75 1.75 70.94 64.94 6.12* 3.59 2.53 11.00 14.49 ferences, independent t-tests of differences Life satisfaction Developmental task between means were computed. In addition, a accomplishment 4.97* 7.56 7.65 50.43 57.19 9.14 one-way analysis of covariance, using health Activity level 8.42 .01 20.25 20.23 and education ratings as covariates and life satisfaction as the criterion measure, was con- ' /K.01, two-tailed test.
Downloaded from http://geronj.oxfordjournals.org/ at UB Giessen on June 29, 2015
tings, for the individuals surveyed from each setting and (2) it afforded a measure of perceived constraint that could be related within each setting to life satisfaction. Each of the six items required a response on a five-point Likert scale from "not at all true" to "completely true." The index was scored in the direction of perceived low situational constraint. The items were: "The rules here keep one from doing the things one wants to do"; "living here makes one dependent upon others"; "Residents are asked for their advice when changes are planned and carried out by administrators"; "Residents have free access to facilities and grounds"; "The staff here responds to residents' requests"; and "If a resident has a problem, he can solve it by taking the initiative." A total score (5-30) was derived. Perceived health and educational level. —For self-perceived health the measure was very similar to that employed by Palmore and Luikart (1972) and Edwards and Klemmack (1973). It consisted of a direct question in regard to perceived health with five alternatives: "Would you say your own health in general is: excellent; good; fair; poor; very poor?" The response to the question was coded from 5 (excellent) to 1 (very poor). The measure of educational level consisted of 7 choices: "My formal education consists of: 1-6 years (elementary school); 7-9 years Gunior high school); 10-11 years (1 or 2 years of high school); 12 years (high school diploma); 13-15 years (several years of college); 16 years (college diploma); 16+ years (graduate school). The response was coded from 1 (1-6 years) to 7 (16+ years).
93
94
WOLKAND TELLEEN The obtained difference between settings for developmental task accomplishment will be discussed in partial support of the role of environmental constraint for over-all adjustment and satisfaction. Indeed, as reported in Table 2, developmental task accomplishment correlated significantly (p < .01) with the degree of perceived autonomy in Setting B. In Setting A perceived health, however, not autonomy, related significantly (p < .05) to developmental task accomplishment. Finally, it is somewhat noteworthy that, although the settings were selected for and did manifest differing levels of psychological and social constraint, the residents of each setting did not report discrepant levels of participation in both formal and informal activities. The data presented in Tables 2 and 3 relate to the investigation of similarities and differences in the correlates of life satisfaction as a function of residential constraint. Table 2 presented the intercorrelation matrix of all variables for each setting. The potential significance of these coefficients resides in an examination of the significant correlates of life satisfaction for each setting. For Setting A these were: perceived health, perceived environmental autonomy, developmental task
Table 2. Correlation Matrix of All Variables by Setting. 3
2
1
Variable
-.24
(1) Education .01
(2) Health
4
5
6
7
-.25
-.20
.21
-.01
-.24
.21
-.02
.48*
.30"
.14
-.08
.31"
.12
.02
-.17
.06
(3) Environmental autonomy
-.10
.08
(4) Self-acceptance
-.18
.12
-.25"
(S) Life satisfaction
.07
.10
.42*
-.38*
(6) Developmental Task
.02
.10
.34*
-.37*
.45*
(7) Activity level
.02
-.12
.07
.04
.16
-.14
.43*
.33" .20
.15
Note: Matrix above the diagonal represents Residence A, that below the diagonal Residence B. V
