| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
a Departments of Internal Medicine and Geriatrics, University of Groningen, The Netherlands
b Department of Psychogerontology, University of Nijmegen, The Netherlands
Nardi Steverink, Departments of Internal Medicine and Geriatrics, University of Groningen, University Hospital, P.O. Box 30.001, NL-9700 RB, Groningen, The Netherlands E-mail: n.steverink{at}int.azg.nl.
Decision Editor: Toni C. Antonucci, PhD
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
THE awareness of age and the experience of growing older are inherent components of the self for most people in their adult years (George, Mutran, and Pennybacker 1980
; Neugarten 1968, Neugarten 1969, Neugarten 1979; Peters 1971; Ryff 1991). As self-conceptions are found to play an important role in behavior and adaptive outcomes (Bandura 1986), this may also be assumed for the cognitions that an individual holds about his or her own aging process (Lerner and Busch-Rossnagel 1981; Markus and Herzog 1992; Ruth and Coleman 1996). In this article, the personal experience of aging is investigated as well as the factors that relate to it, along with the question of whether and how cognitions concerning one's own aging relate to indicators of subjective well-being.
In both literary and philosophical works and in phenomenological studies, the varied nature of the personal aging experience has been extensively described (e.g., Beauvoir 1970; Cicero, as in Baltes and Baltes 1990; Cole and Gadow 1986; Thompson 1992). The empirical investigation of the personal experience of aging, however, is approached almost exclusively from the perspective of subjective age identification or age identity. Subjective age identification refers to just how old a person feels or the age group with which the individual identifies him- or herself (Barak and Stern 1986; Baum and Boxley 1983; Goldsmith and Heiens 1992; Kastenbaum, Derbin, Sabatini, and Artt 1972; Markides and Boldt 1983; Ward 1977). Rather than the personal experience of aging, however, the foregoing research is largely concerned with the personal experience of age.
Also, some studies exist that clearly relate to the concept of the personal experience of aging, but that take a somewhat different perspective. The work of Heckhausen and colleagues, for instance, focuses on personal and normative conceptions about adult developmental change in psychological attributes. It was found that different age groups perceive development across the life span as the coexistence of gains and losses, but also as an increasing risk of decline and a decreasing potential for growth (Heckhausen, Dixon, and Baltes 1989; Heckhausen and Krueger 1993). Another example is the work on possible selves of Markus and colleagues. Possible selves are representations of individuals' ideas of what they might become in the future (Cross and Markus 1991; Markus and Nurius 1986; Ryff 1991). Possible selves are found to change across the life span, which may be interpreted as a component of the personal experience of aging.
Up until now, only a few studies have empirically examined the personal experience of aging framed in terms of aging. For instance, Keller, Leventhal, and Larson 1989 conducted in-depth interviews on the aging experience with 32 community-dwelling adults aged 50 to 80 years and identified five major categories of positive and negative experiences: (a) aging as a natural and gradual process without remarkable features, (b) aging as a period of life evaluation, philosophical reflection, or increased wisdom and maturity, (c) aging as a period of increased freedom, new interests, and fewer demands, (d) aging as a period associated with physical health difficulties or concerns about health, and (e) aging as a period of losses, both interpersonal and job related (Keller et al. 1989). Connidis 1989 also investigated the aging experience of 400 community-dwelling people older than the age of 65 by asking if they liked or disliked anything about being their age and if they had any worries about growing older. These people generally did not focus on the negative aspects of aging. When categorized into three groups of people holding a negative, a moderate, or a positive view of aging, however, those holding a negative view were older, in poorer health, and had fewer children than those holding a positive view. Compared with those holding a moderate view of aging, the people holding a negative view were also older, less financially secure, in poorer health, more likely to have never been married, and more likely to see aging as worse than expected (Connidis 1989).
The present article has three basic aims: first, to investigate the personal experience of aging in such a manner that a wide range of aging conceptions relating to the different domains of the individual's existence are considered; second, to identify those factors that appear to relate to the manner in which the personal process of aging is conceived. Actual age is obviously an important factor, but research has shown that other factors also play a role. Among these are subjective health, being married, income, and the number of children (Connidis 1989). The third and final aim is to investigate whether and how the personal experience of aging—as a cognitive part of the aging self-concept—relates to adaptive outcomes, that is, indicators of subjective well-being. Several studies have shown a positive relationship between age identity (feeling younger) and measures of adjustment among older adults (e.g., George et al. 1980; Montepare and Lachman 1989). Whether a similar relationship holds for the personal experience of aging and measures of adjustment is not as yet clear, however.
Three sets of expectations guided the empirical investigation. First, there are indications that the experience of aging concerns a number of different dimensions of life (Keller et al. 1989) and that it can be framed in both positive and negative terms (Connidis 1989). These findings fit with the life span theoretical perspective on the multidimensionality and multidirectionality of development and aging (Baltes 1987). We therefore hypothesized that the personal experience of aging would reflect different dimensions (be multidimensional), and it would be multidirectional in the sense that it would involve both positive and negative experiences. This kind of multidirectionality we refer to as weak multidirectionality, as opposed to strong multidirectionality, which is considered in the second set of hypotheses, below.
The second set of expectations guiding the present research concerns those factors that may play a role in the origin of personal conceptions of aging. Actual age is obviously an important factor, and strong multidirectionality would imply that, with increasing age, some aging experiences become more positive, whereas others become more negative (Baltes 1987). However, in line with identified age-related changes in conceptions of aging (Connidis 1989; Heckhausen et al. 1989) we hypothesized that strong multidirectionality would not obtain: With increasing age, the experience of aging will become less positive and more negative.
Next to age, other factors relating to individual circumstances or characteristics may also be implicated in the origin of personal conceptions of aging. Relevant factors can be derived from theories that postulate the importance of individual resources (physical and material, social, and psychological resources) for coping with life in general and the aging process in particular (Baltes and Lang 1997; Steverink, Lindenberg, and Ormel 1998). A stable and varied repertoire of resources allows people to not only acquire important goals, but also compensate for the losses that accompany the process of aging. The important physical and material resources for most people in middle and later adulthood are health and financial security: Adapting to the process of aging is generally easier when a person is healthy and without financial worries. The important social resources for most people in middle and later adulthood are having a spouse and children or family and close friends. The important psychological resources for most people in middle and later adulthood are not only their intellectual skills but also their beliefs about self-efficacy and personal agency or feelings of control. Still other resources may exist, but the resources mentioned here are among the most important ones for the majority of adult people. We therefore hypothesized that higher levels of resources will be related to feeling less negative and worried about growing older. That is, the person with more personal resources is better able to cope with life and the process of aging than is the person with less resources. We also hypothesized that the role of resources is stronger for the experience of aging than is actual age. On one hand, actual age is a rather "empty" variable simply reflecting the resources with which it correlates, such as health. On the other hand, actual age may be a very rough indicator of just how much time a person has to live and thereby clearly influences the experience of aging. Given that the amount of time left to live is rather uncertain for most adult people, however, it is hypothesized that the impact of actual age on the experience of aging will be relatively weaker than the impact of resources.
The final set of expectations concerns the generally supported positive relationship between younger age identity and adjustment (e.g., George et al. 1980; Montepare and Lachman 1989). We do not attribute this positive relationship to denial of aging (Montepare and Lachman 1989; Ward 1977), however, but to a complex set of relations involving cognitions, behavior, and adaptive outcomes (Bandura 1986; Markus and Herzog 1992). To reach or maintain a state of well-being across the course of life, individuals must make decisions that enhance their well-being. Decisions of this kind pertain to not only the ability to set goals but also the ability to delay on the rewards of desired outcomes. For instance, regaining the ability to walk after knee surgery requires not only painful effort and exercise, but also the capacity to set a distant goal. The ability to keep such a long-term goal in mind requires not only a delay of gratification but also positive expectations with regard to future outcomes. Whereas a positive expectation (walking again) can further motivate decisions (exercise) that enhance one's well-being, a negative or pessimistic expectation will undermine one's motivation to do the painful exercises needed to walk again. The cognitions concerning one's own aging process can refer to already lived experiences but also to (generalized) expectations with regard to the process of aging in the future. The hypothesis is therefore that people with a positive view or expectation with regard to their aging process will be better able to make the "right" decisions and thus experience higher levels of subjective well-being than will people with a negative view. Put differently, people with a negative view or expectation with regard to their own aging process will be inclined to make the "wrong" decisions and thereby experience low levels of subjective well-being.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
) were between 40 and 85 years of age. They were identified through a national probability sampling of local authorities, both in East and West Germany and were stratified by age, gender, and living region (East or West Germany). The fieldwork was conducted during the first half of 1996. Fifty percent of those contacted (N = 9613) proved willing to participate (n = 4838). The participants were interviewed at home by trained interviewers and asked to complete a "drop-off" questionnaire after the interviewer left. The return rate for the questionnaire was 83.4% (n = 4034). Because most measures used for the present research were collected by means of the drop-off questionnaire, we decided to restrict the analysis to the respondents who completed this questionnaire. Although this final sample still is very large (n = 4034), it consists of less than half of the original sample, which may yield problems of generalizability. Therefore, to identify possible sample selections, we conducted a specific nonresponse analysis (Infas 1997). First, the main characteristics of the people who refused to participate entirely were analyzed. It was found that more people refused to participate because of disability and illness in the highest age group of 70 to 85 years than in the other groups. Also, women generally refused to participate more often than men did (53% vs. 47%) and East Germans were slightly more willing to participate than West Germans were (56% vs. 48%). Second, a comparison was made between the participants who did return the questionnaire after being interviewed (the final sample; n = 4034) and those who did not return the questionnaire (n = 804). Results showed that this nonresponse group, compared with the final sample, contained slightly more women (50.4% vs. 48.8%) and somewhat more West Germans (71.4% vs. 66.1%). There were no differences regarding age. From these analyses it was concluded that, in interpreting the results, it should be kept in mind that the final sample contains a small overrepresentation of healthy people in the oldest age group (70 to 85 years), together with a small overrepresentation of East Germans and men. We return to the point of sample selection and generalizability in the Discussion. In Table 1 , an overview of the main demographic characteristics of the sample can be found.
|
). Examples of the statements are: "Aging means to me ... being less vital and fit" and "Just how well I can take care of myself has nothing to do with my age." Respondents had to indicate whether each statement was "completely true," "mostly true," "mostly not true," or "completely not true." The statements were formulated by members of the research group and were based on the results of two pilot studies in which personal meanings of aging were investigated by means of qualitative methods (Kohli and Dittmann-Kohli 1996; see also Dittmann-Kohli 1995). The resulting initial set of 47 statements was explored using a number of data-reduction techniques. Analyses revealed three convincing clusters of statements regarding the aging experience (see Results).
Physical and material resources.
Two indicators were used to measure physical and material resources: subjective health and income. Subjective health was measured by responding to the question "How do you assess your health at this moment?" with "very good," "good," "moderate," "bad" or "very bad." Income was assessed using equivalence income (Motel 1998) translated into a 14-category classification system with the categories ranging from "less than 1000 German marks per month" (Category 1) to "more than 3400 German marks per month" (Category 14). Information about income was missing in 10.4% of the cases (n = 418).
Social resources.
Three measures were used to assess the availability of social resources. The first measure was "having a spouse," which was dichotomized as "being married or living together with a spouse" versus "being widowed, divorced, or never married." The second measure was the number of children, which ranged from "no children" (score = 0) to "five or more children" (score = 5). For the latter score, the number of children was summed, as only 3.6% of the respondents were found to have five or more children, with 2 respondents having the maximum of 10 children. Twelve percent of the respondents had no children. Finally, a measure of loneliness was added because the other two more objective measures of social resources need not correspond with their subjectively experienced quality. That is, loneliness can reflect the subjectively experienced lack of important social resources and was therefore measured by an 11-item loneliness scale (DeJong-Gierveld and Kamphuis 1985). This scale has good psychometric properties and the internal consistency (Cronbach's 
) for the present study was found to be .89.
Psychological resources.
We used two measures to assess the respondents' psychological resources: level of education and hope. Education was measured with the three categories: low, middle, and high. Hope was measured in terms of personal agency, using an eight-item hope scale (Snyder et al. 1991), to indicate "a cognitive set that is composed of a reciprocally derived sense of successful (a) agency (goal-directed determination) and (b) pathways (planning of ways to meet goals)" (p. 570). For the present study, the internal consistency (Cronbach's ) of this scale was found to be .87.
Subjective well-being.
Subjective well-being was measured by means of both a cognitive (life satisfaction) and two affective (positive and negative affect) components. These are widely used indicators of subjective well-being (Diener 1984; Diener, Suh, Lucas, and Smith 1999). Life satisfaction was measured with the five-item scale, Satisfaction with Life (Pavot and Diener 1993). This scale has good psychometric properties and the internal consistency coefficient for the present study was found to be .86. Positive and negative affect were measured using the Positive and Negative Affect Schedule, which consists of two scales of 10 items each (Watson, Clark, and Tellegen 1988). The internal consistency coefficients for the present study were found to be .87, and .82, respectively, for each scale.
Statistical Procedures
The first set of hypotheses was tested using factor analysis and analysis of internal consistency. The second and third sets of hypotheses were tested using multiple regression analyses. To check for a risk of collinearity in the multiple regression analyses, we examined the intercorrelations between the variables. Inspection of Table 2 shows none of the correlations to exceed .54 or -.43, which can be considered acceptable.
|
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
In light of these findings, we decided to carry out another factor analysis with a three-factor solution. This analysis revealed a first factor with 17 items, a second with 13 items, and a third with 10 items (only items with loadings greater than .40 were considered). The first factor contained negatively formulated items with very diverse content. The second factor, in contrast, contained a diversity of positive items. All of the items on the first and second factors also started with the phrase, "Aging means to me ... ". The third factor contained a mixture of items with 6 of the 10 items ending with the alternative phrase: "... has nothing to do with my age." This finding led us to conclude that the items on the third factor were less convincing, and we decided to continue the analyses with only the items from the first and second factors with loadings of at least .40 (30 items).
As not only weak multidirectionality (positive and negative experiences), but also multidimensionality were expected to characterize the personal experience of aging, we decided to subject the 30 items selected on the basis of the previous analysis to another factor analysis. The result was five factors with eigenvalues of 7.1, 3.6, 1.9, 1.2, and 1.1, respectively, explaining 49.6% of the variance when considered together. The first factor contained seven items (with loadings >.40); all of the items referred to the aging experience of physical decline and problems of coping, with the exception of one item, which was then removed. The second factor also contained seven items (>.40) reflecting the positive aging experience of continued personal development, expansion, and new experiences. The third factor contained four items (>.40) referring to losses in the social domain. The fourth factor revealed three items (>.40) with the positive experiences of compensation ability, control, and self-knowledge. Finally, the fifth factor contained six varied items referring to alienation, negative personality characteristics, dissatisfaction, and curiosity about the future.
To determine whether each factor represented a meaningful and coherent dimension of the aging experience, to use as separate scales in the further analyses, we further analyzed each factor with respect to internal consistency. An internal consistency coefficient (Cronbach's ) of at least .70 and meaningful content were taken as the criteria for acceptance.
The results of this analysis revealed the following. The six items from the first factor showed an internal consistency coefficient (Cronbach's ) of .82. The seven items from the second factor showed an internal consistency coefficient of .81. The four items from the third factor revealed a Cronbach's alpha of .77 and the three items from the fourth factor and the six items from the fifth factor revealed coefficients of .69 and .65, respectively.
According to the criteria, the fourth and fifth factors were unacceptable. The internal consistency of the fourth factor could not be increased because this factor had only three items. It was therefore decided to remove the items. Similarly, deletion of one or two of the items from the fifth factor did not increase the internal consistency of this factor, so it was also decided to remove these items. This led to a final set of three factors.
Before performing further analyses, we decided to delete two more items from the first factor. Both of the items loaded considerably smaller than the other items on this factor, and one of the items was considered vague with regard to content (it focused on coping with life in general). The resulting factor or dimension could be labeled as aging experiences of physical decline. An additional item was similarly deleted from the second factor because it showed a somewhat weaker loading than the other items. This factor or dimension could be labeled as aging experiences of continuous growth. Inspection of the third factor showed no reason for change, so the four items loading on it were retained. The third factor or dimension could be labeled as aging experiences of social loss.
It was next observed that the first and third factors each contained four items, whereas the second factor contained six items. For purposes of comparison, we deemed it desirable to have the same number of items on each factor, and we therefore considered restriction of the second factor to four items. Reduction of the second factor to four items did not appear to damage the meaningfulness or internal consistency of the factor; only a slight change in the internal consistency of the items was observed after the omission of two items (from = .81 to = .78). In Table 3 , the final results of the factor analysis and the internal consistency of the items are summarized.
|
) for the three factors were found to be .79, .78, and .77, which can be considered satisfactory. The conclusion is that the subjective experience of aging is both multidimensional and multidirectional, with the latter implying the coexistence of both negative and positive experiences (i.e., weak multidirectionality). The multidimensionality of the aging experience can be seen to pertain to the physical-, social-, and personal-development domains of life.
In the second set of hypotheses we stated, first, that age would have an influence on the personal experience of aging such that strong multidirectionality would not obtain: With increasing age, the experience of aging becomes less positive and more negative. The results regarding this hypothesis show the following. All three factors or dimensions of the aging experience are found to correlate significantly with age in the expected direction (p < .01). Whereas the dimensions of physical decline and continuous growth produced age correlations of r = .31 and r = -.35, respectively, the dimension of social loss revealed an age correlation of r = .18. To check the linearity of these relationships, we compared the mean scores for the six age groups (40–46, 47–54, 55–61, 62–69, 70–76, and 77–85 years) on each dimension. A clear linear relationship with age was found for each of the three dimensions of the personal experience of aging (weighted cases). Both the negative dimensions (physical decline and social loss) show that the older one is, the more the aging experience is framed either in terms of physical decline, F(3895,5) = 83.35, p < .001, or in terms of social loss, F(3944,5) = 35.58, p < .001. The dimension of continuous growth shows the process of aging to be experienced as less of a process of continuous growth with increasing age, F(3846,5) = 124.01, p < .001.
In Table 4 , the results pertaining to the other hypotheses of the second set of hypotheses are shown. It was stated that individuals with higher levels of resources could be expected to be generally less negative and feel more positive about growing older than would individuals with lower levels of resources. Also, although actual age is expected to have an independent influence on the personal experience of aging, the influence of resources was expected to be greater. We performed separate regression analyses with the three dimensions of the personal experience of aging as the dependent variables. All three models are shown in Table 4 , each model containing two submodels. Each first submodel includes only age; each second submodel also includes the following resources: subjective health, income, having a spouse, number of children, loneliness, education, and hope. Also, gender and living region in Germany (West or East) were added to each second submodel to allow us to examine the possible influence of these variables.
|
The resource of education also relates differentially to the three dimensions of the aging experience. A higher educational level relates to the experience of aging as less of a social loss, and more as an opportunity for continuous growth. The relation to physical decline was nonsignificant, however. A higher educational level does not, thus, seem to protect an individual from experiencing aging as physical decline. Finally, one of the resources considered here—number of children—did not influence any dimension of the aging experience, which is certainly unexpected and is considered in further detail in the Discussion.
Comparison of the standardized regression coefficients (ß) for each second model shows only a few of the resources playing a stronger role in how people experience the aging process than actual age. The aging experience of physical decline appears to be better predicted by subjective health than by actual age. The aging experience of continuous growth appears to be better predicted by hope than by actual age. Additionally, the aging experience of social loss appears to be better predicted by loneliness and hope than by actual age. These findings lead us to conclude that age is a relatively strong and independent determinant of just how the process of aging is experienced. Only a few resources (or lack of them) appear to affect the aging experience more than actual age.
With respect to gender, it is remarkable that it does not relate to the experience of aging as physical decline but does relate to the experience of aging as both social loss and an opportunity for continuous growth. Men and women apparently do not differ with respect to the experience of aging as physical decline, whereas men tend to feel greater social loss as a result of aging than women, and women experience more continuous growth than men.
With respect to region of living, it is interesting to note that East Germans seem to experience the aging process as less of an opportunity for continuous growth than do West Germans, although they do not differ with respect to the experiences of physical decline and social loss. This finding is hard to interpret but, in light of the fact that the influences of income and health have been taken into consideration, it may simply indicate a more general lack of opportunities for continuous growth in East Germany relative to West Germany.
The third set of hypotheses in the present study concerned the influence of the different dimensions of the personal experience of aging on measures of subjective well-being (i.e., life satisfaction, positive affect, and negative affect). Note here that, although the rationale for the relationship between (different dimensions of) the aging experience and indicators of subjective well-being is based on the hypothesized intermediate mechanism of "making the right decisions" (linking positive cognitions to adaptive outcomes and negative cognitions to maladaptive outcomes), this intermediate step could not be measured with the available data. Nevertheless, we deemed it acceptable to interpret the empirically found relationships—if supported—as such.
The results of separate regression analyses with the three measures of subjective well-being as the dependent variables are presented in Table 5 . Again, each model consists of two submodels. Each first submodel contains age, resources, gender, and region as independent variables. Each second submodel also includes the three dimensions of the personal aging experience.
|
In general, it may be concluded that the third set of hypotheses is largely supported by the data. The way in which people experience their own aging process independently affects how they feel in terms of both life satisfaction and positive or negative affect. Comparison of the standardized coefficients (ß) for the three dimensions of the personal aging experience together with the increase in the amount of explained variance, however, shows the impact of various aspects of the experience on life satisfaction to be very small, the impact on positive affect to be particularly due to the experience of continuous growth, and the impact on negative affect to be mainly a matter of physical decline and social loss.
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
The findings primarily showed the personal experience of aging to be a multidimensional phenomenon. Three different dimensions of aging experiences were found. The first dimension relates to physical decline, such as the loss of vitality and health. The second dimension relates to continuous growth and personal development. The third dimension relates to losses in the social domain, such as no longer being needed by others or decreased respect from others. The three dimensions of the aging experience found here are largely in line with the findings of a study by Keller and colleagues 1989, who identified two additional dimensions: Aging as a natural and gradual process with little or no remarkable features and aging as a period of life evaluation, philosophical reflection, and increased wisdom. Although our data initially also revealed more than three dimensions (see Results), only three dimensions could be interpreted unambiguously while at the same being internally consistent scales. The fact that Keller and colleagues 1989 reported on more than three dimensions may lie in their use of open-ended questions to tap the subjective experience of aging. Such a qualitative procedure (but only in 32 adults aged 50 to 80) may yield results that are hard to replicate in a large quantitative and statistically controlled study. Conversely, the set of statements used in our study may have been insufficient in content and/or number to identify other statistically significant dimensions. Further research should take these points into consideration.
A second finding concerns the multidirectionality of the aging experience, which, on the one hand, is reflected by the fact that both positive (growth) and negative (loss and decline) experiences characterize the personal experience of aging (i.e., weak multidirectionality). On the other hand, we found, as expected, that strong multidirectionality of aging experiences does not obtain: With increasing age, the aging experience is increasingly framed in terms of physical decline or social loss and less in terms of continuous growth.
A third major finding in the present study is that, next to age, a substantial number of the resources were significantly related to each of the three dimensions of the aging experience. Adults with better subjective health, a higher income, less loneliness, a higher educational level, and stronger hope experience the process of aging less in terms of physical decline and social losses and more in terms of continuous growth. Only two of the resources examined in our research did not relate to the experience of aging in the manner expected. People with fewer children were found to hold a more negative view of aging than were people with more children in a study by Connidis 1989, whereas our data showed no differences. One explanation for this may lie in the fact that our study covered the range of 40 to 85 years, whereas the study by Connidis involved only adults over the age of 65 years. Younger participants are possibly less aware of how dependent they can become on their children when they age. Nevertheless, it may also be the case that it is not so much the number of children that is important for feeling more or less worried about one's own aging process, but having any children (versus having no children). Moreover, having children as a continuous variable may contain the risk of curvilinearity. To check these possibilities, we also executed all regression analyses with having children as a dichotomous variable. This, however, did not yield other results. Another possible explanation may still be that it is not so much the number of children that is important, but the quality of the relationship with one's children. As we also considered the perceived quality of one's social relations (including children) indirectly through the level of loneliness, the possible influence of number of children may have been masked by the significant relationship of loneliness to all three dimensions of the aging experience.
The other resource not relating to the experience of aging in the manner expected was having a spouse. In the study by Connidis 1989, the people holding a negative view of aging were more likely to have never been married; in our study, the people less likely to experience aging as a social loss were married. However, having (or not having) a spouse did not relate to experiencing the process of aging in terms of physical decline or continuous growth. It seems that the implications of having (or not having) a spouse for one's experience of aging primarily refer to the social domain. This finding can be interpreted as a differentiation of the finding of Connidis 1989 and shows the importance of investigating the aging experience not only in positive or negative terms, but also in different dimensions.
In addition to several resources, actual age was found to influence the three dimensions of the aging experience rather strongly, which was not quite as expected. The influence of age on the different dimensions of the aging experience appeared to be sometimes equal but often stronger than most of the resources examined. The higher the age, the more inclined the people were to frame the aging process in terms of physical decline and social loss and less in terms of continuous growth, regardless of most of their resources. Only three resources appeared to have a stronger impact on the dimensions of the aging experience than actual age. First, subjective health influenced the aging experience of physical decline more than actual age did. This means that actually feeling less healthy and not so much being older leads one to experience the process of aging in terms of physical decline. Also, the experience of aging in terms of continuous growth appears to depend more on a higher level of hope than on being younger. Finally, the experience of aging in terms of social loss seems to be triggered by loneliness and a low level of hope rather than by simply being older. The findings regarding these three resources partly support our hypothesis that the impact of resources on the personal aging experience would be stronger than actual age. Besides, it shows that evaluation of these three resources in particular is associated with the personal process of aging. Here it should be noted also that there may be a difference between actual resources such as income, spouse, and education, and perceived resources such as subjective health, loneliness, and hope, when relating them to the different dimensions of the aging experience. Perceived resources may have a different relationship to the three dimensions of aging experiences than do actual resources, because both perceived resources and aging experiences are measured as subjective evaluations (i.e., perception), whereas actual resources are measured as objective facts. What remains to be seen, however, is whether such evaluations are based on real experiences or on misconceptions about aging. For instance, people can be lonely because they lack companionship and not because of their age. Still, they may be inclined to attribute this experience to the aging process.
Also the finding that age still exerts a relatively strong independent influence on the personal experience of aging after all resources (including health, loneliness, and hope) were controlled for is intriguing. To examine whether stepwise multiple regression analyses would yield more insight into the separate impact of age and the different resources, we also executed all regression models using stepwise methods. This, however, did not change the results. The relatively strong influence of age may also suggest that, although they were asked about their personal experience of aging, people may still be inclined to intermingle personal experiences of aging with normative (i.e., more negative and less positive) expectations. Normative expectations are associated with actual age. Therefore, even when the influence of other important factors (i.e., resources) has been controlled for, actual age may still have a relatively strong impact on the personal experience of aging through internalized normative expectations with regard to aging. These findings are intriguing and in need of further research, particularly when it is realized that misconceptions about aging may lead to unnecessary worries about the process and may ultimately have a negative impact on the quality of life (see also Neikrug 1998).
Our results regarding the relations between the personal experience of aging and measures of subjective well-being (i.e., life satisfaction, positive affect, negative affect) also point to the negative effect of misconceptions about aging on quality of life. All three measures of subjective well-being were found to be substantially influenced by at least two of the three dimensions of the aging experience. Only two effects deviated from the expected pattern of results. First, experiences of physical decline do not appear to influence life satisfaction although they do relate to both positive and negative affect. This finding may be due to the fact that life satisfaction was measured in terms of overall satisfaction with one's life. Moreover, experiences of physical decline may be seen as trivial and generally accepted as a normal part of the aging process. Positive and negative affect, however, are more state measures and thus, perhaps, more sensitive to feelings of physical decline. Nevertheless, it remains intriguing that experiences of social loss and continuous growth, which indeed relate significantly to life satisfaction, apparently are not considered trivial or simply part of the aging process. The other unexpected finding was the positive relationship between the experience of aging as continuous growth and feelings of negative affect. This may be due in part to the positive correlation between positive and negative affect (r = .14, see Table 2 ). Another explanation, however, is that people who experience aging in terms of continuous growth are open to new opportunities, which may sometimes yield positive experiences and sometimes negative ones. Further research is needed to elucidate these findings.
Three final points for discussion and future research should be mentioned. The first concerns the relatively large nonresponse this study encountered and its consequences for interpreting the results. Although such a rate of nonresponse is common in large-scale population surveys, it urges researchers to be cautious when generalizing the results to the German population. On the other hand it may be argued that, because the main sample biases could be identified (see Methods section), it is known at what points the generalizability is limited. Moreover, as this article also aimed at elaborating theoretical issues and model building, the study is deemed interesting for that purpose as well.
The second point concerns causality. The direction of causality for a number of the relations observed in the present, cross-sectional, research can only be determined on the basis of longitudinal data. Nevertheless, as a first attempt to shed some light on the personal experience of aging and the factors particularly relevant to it, a cross-sectional design can provide some useful insights, particularly when the analyses of the cross-sectional data are guided by specific theoretical considerations. Moreover, some useful insights for further (longitudinal) examination may be gained in such a manner.
The last point for consideration in future research is whether other resources or particular personality characteristics such as optimism may play a role in people's cognitions about their own aging process. The substantial amount of variance left unexplained in the present analyses suggest such a possibility.
It may be concluded that the personal experience of aging encompasses more than just a particular age identity. The personal experience of aging has a number of different dimensions and includes both negative and positive experiences. Furthermore, the personal experience of aging seems to play a role in just how happy people may feel in terms of the degrees of both positive and negative affect. The present findings may contribute to our knowledge of the personal experience of aging and may help identify those people at risk for developing a particular negative view of the process of aging. Such information can also be used to develop interventions aimed at a more positive aging experience and greater subjective well-being in later life.
|
Acknowledgments |
|---|
Received for publication September 9, 1999. Accepted for publication July 31, 2000.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. J. Westerhof and C. L. M. Keyes After the fall of the berlin wall: perceptions and consequences of stability and change among middle-aged and older East and west germans. J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2006; 61(5): S240 - S247. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. R. Lang, N. Rieckmann, and M. M. Baltes Adapting to Aging Losses: Do Resources Facilitate Strategies of Selection, Compensation, and Optimization in Everyday Functioning? J. Gerontol. B. Psychol. Sci. Soc. Sci., November 1, 2002; 57(6): P501 - 509. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
