| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
1 University Honors College, Oregon State University, Corvallis.
2 Department of Sociology and Center for the Study of Aging, University of Vermont, Burlington.
Address correspondence to Dr. Jon Hendricks, University Honors College, Oregon State University, Corvallis, OR 97331. E-mail: hendricj{at}onid.orst.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Total number of organizations volunteered for and total number of hours engaged in volunteer activities were utilized to obtain measures of volunteer hours per organization and volunteer hours in the main organization to determine whether a selective process could be observed. Descriptive statistics on age patterns were followed by a series of curve estimations to identify the best-fitting curves.
Results. Logistic age patterns of slowly increasing then relatively stable volunteer activity suggest that socioemotional selectivity processes are operative in the realm of voluntary activities.
Discussion. Socioemotional selectivity theory is applicable to voluntary activities.
An abiding question in social gerontology is often phrased in terms of whether it is better to maintain middle-age precedents of activity and engagement or to relinquish such patterns and go gently into old age. Two less-than-parallel perspectives have shaped much of the dialogue over the last four decades. One view maintains that given the salutary effects of nearly all social interaction, a person moving into the later years would be well advised to maximize interaction and involvement, surrendering as little as possible as late as possible (Janoski & Wilson, 1995
; Monk, 1995; Musick, Herzog, & House, 1999; Rotolo, 1999; Rowe & Kahn, 1998). Some investigators even assert that a "drive to engage" remains viable until quite late in life, promoting productive involvements until well into old age (Weiss & Bass, 2002). The antithetical point of view is that declines are natural, inevitable, even functional, so it is better not to try to stem the tide but to go with the flow and slowly back away (Baltes, 1996; Baltes & Smith, 1999). Of course, both views are more highly nuanced and dynamic than the above caricature; still, it is a reasonable representation of activity and continuity theory, on the one hand, and disengagement theory, on the other (Carstensen, 1992).
The obvious riposte to such a dichotomy is that those cannot be the only two choices. In a quest for a middle ground, Carstensen (1993) articulated what she termed "socioemotional selectivity theory" and, together with a number of colleagues, has sought empirical validation of a lifelong process of judicious connections. In a nutshell, the theory maintains that through selective optimization, losses and rescissions in a person's life are countered by increasing investments in and the relative salience of other agenda items. Carstensen is of the opinion that a strategy of relative involvement may be thought of as a means of maintaining a sense of well-being in the face of loss or finitude (Carstensen, 1993, 1995). This investigation is an attempt to ascertain if the dynamic is evident in the realm of voluntarism. We utilize a substantial national data set to examine differences in the breadth and intensity of voluntary participation as a selective practice.
Socioemotional Selectivity Theory
As Carstensen (1987, 1991, 1992, 1995) has noted in a series of landmark publications, many available opportunities seemingly go underutilized among older people. Why? She hypothesized that beginning early in life, people adopt a deliberate strategy to decide on which opportunities to pursue and which to let pass. Those that persist may evolve over time, but their functional relevance remains more or less stable. Certain engagements or options are avoided or relinquished because actors are risk averse or because they choose to maximize return on their investments of time, energy, or emotionality. In later life, the process becomes particularly apparent, however, out of an awareness of limitations of time and energy (Carstensen, 1992). A related aspect of actors' selectivity concerns what has been termed "self-relevancy" (Hendricks, 1999; Marcus & Cross, 1990; Whitbourne & Primus, 1996). That is, social interaction and events are always meaning-making, virtually by definition, and that fact is apparent in life course patterns of participation, behavior, and social contacts. Interaction and involvements, even interpretations, that are linked to self-concepts and identity persist; those that do not are attenuated or supplanted. Carstensen (1992) goes further, asserting that new activities or interaction with unfamiliar people may yield diminishing returns vis-à-vis self-referential processes and are therefore judged as not worth the effort. In effect, those areas that generate the greatest potential affective or practical return receive attention and those of lesser relevance in those metrics pass from the scene. She points out that the relative significance of all manner of involvements changes over the life course, reflecting differential priorities of the moment, plus construal of what the future portends (Carstensen, 1993, 1995). In a word, preferences change, and furthermore, the dynamic prompting that change reflects important socioemotional priorities embedded in the aging process (Baltes, 1987; Baltes & Baltes, 1990; Carstensen, 1992, 1993, 1995).
In a succession of empirical investigations, Carstensen and colleagues found that some types of participation persist or even increase, while other more peripheral relationships and categories of participation decline as actors exercise a strategy of selective optimization (Carstensen, 1987, 1991, 1995; Fredrickson & Carstensen, 1990). This conservation principle implies that actors concentrate their investments—that over time, their social radius may recede, but the salience of what remains will intensify. Carstensen and others are unequivocal: Focusing on a favored range of interactions or experiences is not a negative accouterment of aging but a natural process that makes all the sense in the world in terms of maximizing emotional gratification (Baltes & Baltes, 1990; Baltes & Smith, 1999; Carstensen, 1992; Okun & Schultz, 2003).
Utilizing a nationally representative sample, Lansford, Sherman, and Antonucci (1998) sought to verify Carstensen's basic notion by looking at age differences in contacts with social networks. The authors found that declines in network size did not presage dissatisfaction and that individuals were happy, by and large, with networks regardless of their relative size. Interestingly, there were no age-related differences in absolute network size observed by Lansford and colleagues, and they concluded that their results do not support the suggestion of selective cut-backs in social contacts as far as they could discern, based on retrospective reports across a 5-year period. In an investigation of social participation among older widowed and nonwidowed individuals, Utz, Carr, Nesse, and Wortman (2002) commented that their analysis was unable to determine whether a selective contraction of forms of social engagement occurs prior to age 65, though they did find that informal participation held steady while formal participation declined after that age in both study groups. They do, however, posit some degree of support in their findings for Carstensen's socioemotional selectivity theory. These tantalizing but less-than-congruent findings helped focus the current investigation into patterns of volunteerism among older persons.
Volunteerism and Age
Volunteerism is widely recognized as an important source of satisfaction, sociability, and self-validation over the life course (Pillemer, Moen, Wethington, & Glasgow, 2000; Ray, 2002; Rowe & Kahn, 1998). Contrary to assertions about Americans losing sight of collectivism (Putnam, 1995, 2000; Rotolo, 1999), rates of voluntarism among the elderly, in particular, likely do not fall as far or as precipitously as the lore suggests (American Association of Retired Persons [AARP], 2003; Chambré, 1993; Fischer & Schaffer, 1993; Goss, 1999; Monk, 1995; Smith, 1994). It has long been asserted that there is a life-stage relationship to volunteering attributable to linkages with family and work, and those rates are low on either side of a middle-aged peak (Choi, 2003; Independent Sector, 2001). Others maintain that the data are inconsistent, and some research asserts quite the converse may be the case (Cutler & Hendricks, 2000; Hendricks & Cutler, 2001).
Insofar as the percentages of volunteering among successive cohorts may trend downward, the presumption of declining rates has fueled a conventional claim that older persons withdraw from voluntary participation (Burr, Caro, & Moorhead, 2002; Cutler, 1976; Independent Sector, 2001). More systematic analyses have questioned this one-dimensional claim by pointing out that those compositional variables characteristic of older cohorts are exactly those also known to be associated with low rates of volunteering. In other words, controlling for compositional variables removes a substantial portion of the apparent decline (Cutler & Hendricks, 2000). Indeed, we assert that insofar as volunteerism is a significant source of subjectively salient involvement, that fact is unlikely to change appreciably with age. Regardless of rates, investigators speak with virtually a unified voice in averring that volunteerism is meaningful for older adults on any number of levels and for a multitude of reasons. For individuals, volunteering provides a sense of well-being, of making valuable contributions to society, of paying something back, of being able to compensate or substitute volunteer roles for others that may have fallen away, and for maintaining a sense of being part of an on-going agenda (Cutler & Hendricks, 2000; Morrow-Howell, Hinterlong, Rozario, & Tang, 2003; Morrow-Howell, Hinterlong, & Sherradan, 2001; Musick et al., 1999; Pillemer et al., 2000; Shmotkin, Blumstein, & Modan, 2003; Van Willigen, 2000).
In attempting to explain relative rates of volunteerism, it has long been asserted that voluntary participation is closely aligned with other dynamic engagements and will ebb and flow accordingly. The family life cycle is perhaps a prime case in point that has been utilized to explain rates and types of voluntary participation (Knoke & Thompson, 1977; Rotolo, 1999), though other life course events have also received attention (Cutler, 1976; Moen, 1996; Musick et al., 1999; Utz et al., 2002). Generally speaking, the contention has been that successive life course transitions will bring realignments in forms and rates of voluntary participation (Rotolo, 2000). Investigations into the trajectory of voluntary association activity have not yet addressed in any systematic way the possibility that socioemotional selectivity may play a role, focusing instead on triggers thought to be associated with life course events. Clearly, the two are parallel processes and may confound interpretation of why age differences appear as they do. Event history analysis concentrating on role shifts may actually mask processes associated with selective optimization.
Fung, Carstensen, and Lang (2001) provide a valuable impetus for examining social participation in their assertion that engagement in peripheral activities declines in favor of more "expressive" and emotionally meaningful forms of engagement as people become aware that their futures are finite. In a word, network size and forms of involvement shrink as people exercise judicious choices about where to focus their energies. A clear age-based pattern is implied: Older people will have fewer forms of engagement than their younger counterparts and will retain more familiar forms of involvement and those that speak to their own self-referential agenda. In effect, a centripetal process may operate, but sense of well-being and satisfaction with what remains stays constant or increases. Whether this centripetal process implied by socioemotional selectivity theory applies to the realm of volunteerism is the principal question addressed by this study.
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
, for additional detail). Although information about the household and each of its members is usually collected via in-person or telephone interviews from a single knowledgeable respondent, the 2002 volunteer supplement attempted to include self-responses from all household members 15 years and older. Fielded September 15–21, 2002, the supplement asks about voluntary activities during the year prior to the interview. General areas covered included frequency of volunteerism, types of organizations in which respondents volunteered, and kinds of activities performed. Persons who did not volunteer were asked what kept them from volunteering and what would encourage them to do so.
Sample Size
The September 2002 CPS includes data on a total of 143,569 persons residing in 56,787 households. Of these, information on whether persons have done any volunteer activities in the last year, collected about people age 15 and older, is available for 97,311 individuals residing in 50,796 households. Our principal analyses, on volunteers who compose 29.6% of those reporting on their volunteer status, are based on responses from 28,757 persons residing in 15,863 households.
Measures
Because a central objective is to examine age patterns of engaging in volunteer activities, we have recoded age into 14 5-year intervals, from 15–19 years of age to 80 and older. Although sample size would permit even further differentiation among persons 80 years and older (N = 4,105), beginning in March 2002, age has been top-coded in the CPS at 80 rather than 90, as had been the case previously, owing to considerations of confidentiality in public use data files.
The basic measure of having engaged in a volunteer activity is derived from responses to two questions. Respondents were first asked, "Since September 1 of last year, have you done any volunteer activities through or for an organization?" Persons who responded "no" or "don't know" were asked a follow-up question: "Sometimes people don't think of activities they do infrequently or activities they do for children's schools or youth organizations as volunteer activities. Since September 1 of last year, have you done any of these types of activities?" Combining the responses from the two questions yields a distribution of 29.6% who have engaged in volunteer activities in the last year and 70.4% who have not by their own recounting.
We examine the age patterning of total levels of volunteer activities with three measures. To determine the number of organizations in which volunteer activities occurred, respondents were asked, "How many different organizations have you volunteered through or for in the last year, that is, since September 1, 2001?" This information was utilized to construct both a dichotomous indicator of whether the respondent volunteered in the last year as well as a count of the number of organizations in which volunteer activity occurred. For each organization mentioned, information on the total number of hours spent volunteering was also collected, and these were summed across organizations to yield a measure of the total number of hours persons engaged in volunteer activities during the last year.
Socioemotional selectivity theory, however, shifts the focus away from age patterns in total number of organizations or total number of hours volunteered to measures that more nearly reflect the shedding of peripheral volunteer activities and retention of core activities. Two measures available from the CPS data appear to capture this emphasis on maintaining core types of engagement. First, recognizing that older persons may belong to fewer organizations on average or devote fewer total hours to volunteer activities, we constructed a measure of the average number of hours spent per organization, which, in effect, controls for the number of organizations in or through which people participate. Second, the CPS asked a series of questions about participation in the organization in which the volunteer has spent the most time during the last year (e.g., nature of the volunteer activities, how the person first became a volunteer). Included among this information is the number of hours the person volunteered for this "main" organization. Consistent with the postulates of socioemotional selectivity theory, the number of volunteer hours devoted to central or main organizations would be expected to remain stable with increasing age, even though number of volunteer hours devoted to all organizations may diminish.
Descriptive statistics on each of the measures of volunteer activity are presented in Table 1.
|
In the second stage of the analysis, we provide a more rigorous test of the applicability of socioemotional selectivity theory to volunteering. We do so by estimating a series of curves describing the relationships (a) between age (in single years) and average number of hours volunteered per organization and (b) between age and total hours of volunteer activities in the main organizations. If the predictions following from the theory are applicable to voluntary activity, the best-fitting curve should be neither linear nor curvilinear, but rather a logistic curve that rises through the middle years and then remains relatively constant at the older ages. We rely on the curve estimation procedure in SPSS (version 11; Chicago, IL) to generate the data needed to select the best-fitting curve. This procedure provides regression statistics for a variety of curve estimation models, allowing one to evaluate the goodness of fit of alternative models to the data.
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Interestingly, the pattern of age group differences in total hours volunteered during the last year departs considerably from the curvilinear patterns seen for the two previous measures. Although there are some irregularities, the total number of hours volunteered generally increases in linear fashion through ages 75–79, followed by a notable decrease only for the oldest age group. These data would seem to suggest that the retirement years provide an opportunity for a greater investment of time in voluntary activities. Owing to the top-coding of the age variable, it is not reasonable to assume that declines occur at any particular point, only that they do taper off in the years after age 80.
The critical test of the postulates of socioemotional selectivity theory comes with our final two measures. If there is a tendency among older persons to shed peripheral volunteer activities and retain core involvements, the average number of hours of volunteer activities per organization and the total number of hours spent in volunteer activities in the main organization should show no evidence of a curvilinear pattern by age. Although some irregularities between adjacent age intervals again appear, the only evidence of a decline is at the very oldest ages. Even there, the average number of hours volunteered per organization over the last year among persons 80+ is higher than for every other age group except for those 75–79. Similarly, the total number of hours people 80+ spent volunteering in their main organization is exceeded only by the hours persons 70–74 and 75–79 years of age devoted to their main organization during the last year. For ease of visual inspection, these relationships between age and average number of hours of volunteer activity per organization and between age and total number of hours spent in volunteer activities in the main organization are plotted in Figure 1.
|
Four models are estimated to assess the best fit. These include (a) a linear model (Y = a + bX), predicting an increase with age in average number of hours spent volunteering per organization and in the total number of hours spent in volunteer activities in the main organization; (b) a quadratic model (Y = a + bx2), which predicts a curvilinear relationship peaking in the middle years and declining at the older ages; (c) a logarithmic model (Y = a + b log X), predicting an increase in volunteering with age but with a declining rate of increase at the older ages; and finally (d) a logistic model (Y = 1/[1 + ae–bX]), where volunteer activity increases with age up to a point after which the increase diminishes and stabilizes. Hypothetical examples of each of these curves are presented in Figure 2, and summary regression statistics for each of the actual models are given in Table 3.
|
|
). Although our focus has been on the two measures of voluntary activity that are unusually well suited to testing socioemotional selectivity theory, another tenet of the theory is, of course, the shedding of organizations in later life. This should be reflected in a curvilinear relationship between age and number of organizations volunteered for. Of the four models tested, the data (not presented here) do indeed show that the best fit is provided by a quadratic model and that the difference in R2 between it and the next best-fitting model (a logarithmic model) is statistically significant.
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
). According to the theory, there is a lifelong selection process wherein people make strategic investments in certain self-relevant or emotionally meaningful goals. With the passing of the years, interaction and activities that yield positive emotions are sustained, while more peripheral activities are jettisoned as not worth an investment of time or energy. Any pattern observed among older persons is interpreted as the upshot of this lifelong process and represents a harmonizing of interests with energies (Carstensen, 1992). As Carstensen and colleagues aver, strong bonds are retained over time among older persons, but fewer peripheral partners and activities are retained than among younger individuals (Fung et al., 2001). Hence, recognition of limited horizons brings about reasonable adaptive changes whereby less salient linkages are abandoned in favor of those with the greatest subjective meaning.
In this analysis, we have focused on selected measures of volunteer activity that are particularly well suited to testing the principal postulates of socioemotional selectivity theory. As our analysis implies, not all organizations in which older people volunteer are alike (Cutler & Danigelis, 1993), so it should not be surprising that involvement in some is more meaningful than in others. The fact that this cross-sectional look at voluntary participation among a nationally representative sample of some 28,000 individuals identified logistic curves suggesting long slow patterns of increasing volunteer activity with age that then remain impressively constant lends immediate credence to socioemotional selectivity theory. On the basis of these data, one might reasonably conclude that there is indeed a type of selectivity operating in the realm of voluntary activities. Fung and Carstensen (2004) have recently begun to contrast alternative explanations to socioemotional optimization in the face of limited time and resources, but to date they are not persuaded that the original formulation is seriously undermined.
Our results also suggest that older volunteers have the potential of contributing even more to societal well-being than is sometimes thought to be the case. The analysis reported here is based on volunteer activities undertaken through or for an organization. Further research on the fit between socioemotional selectivity theory and volunteering might use the type of an expanded characterization of volunteering employed in a recent AARP-sponsored survey of broadly defined voluntary activities. Redefining measures of productivity to include voluntary contributions in formal as well as in informal settings will surely counter the notion that older people do not pull their own weight (Hendricks, 2001; Morrow-Howell et al., 2001). According to information collected by AARP (2003), some 51% of middle-aged and older persons report one or another type of formal volunteerism, with an additional 36% providing informal voluntary contributions not necessarily recognized by traditional questions.
As suggestive as our results are, longitudinal data permitting multivariate analysis would have been preferable for purposes of confirmation. Nonetheless, the robustness of these data is reassuring that the patterns identified are indeed plausible and deserving of further inquiry. As our R values imply, however, there are any number of factors in addition to age influencing volunteering. The task remains for multivariate research to identify these factors and to sort out their relative importance for both formal and informal forms of volunteering. Finally, it must also be mentioned that the top-coding of all respondents over the age of 80 in the CPS dataset into a single category may mask the ultimate pattern of voluntary participation even while preserving anonymity of respondents. Use of a higher age would have been helpful, say 85+, to at least conform to commonly drawn distinctions by age among older persons and to allow a focus on the oldest old.
Assuming the credibility of these data, the results are suggestive of a possible trend toward higher rates of volunteerism among older persons. Although no such conclusion can be asserted absent replicated measures assessing participation rates, these results do point toward a policy promoting more fitting options for better utilization of older volunteers and more suitable targeting of solicitations for volunteers. If older volunteers do make judicious choices based on the most salient forms of participation, organizations and causes would be well served by highlighting the meaningfulness of volunteer opportunities and more carefully placing those volunteers who do step forward. With improved matching of opportunities and interests, older volunteers will continue to make valuable contributions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication January 8, 2004. Accepted for publication April 27, 2004.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. C. Franklin and C. A. Tate Lifestyle and Successful Aging: An Overview American Journal of Lifestyle Medicine, January 1, 2009; 3(1): 6 - 11. [Abstract] [PDF]
|
|
|
|
 |
|
 B. Kaskie, S. Imhof, J. Cavanaugh, and K. Culp Civic Engagement as a Retirement Role for Aging Americans Gerontologist, June 1, 2008; 48(3): 368 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Fengyan Tang Socioeconomic Disparities in Voluntary Organization Involvement Among Older Adults Nonprofit and Voluntary Sector Quarterly, March 1, 2008; 37(1): 57 - 75. [Abstract] [PDF]
|
|
|
|
|
|
T. D. Windsor, K. J. Anstey, and B. Rodgers Volunteering and Psychological Well-Being Among Young-Old Adults: How Much Is Too Much? Gerontologist, February 1, 2008; 48(1): 59 - 70. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Dunn and C. K. Riley-Doucet Self-Care Activities Captured Through Discussion Among Community-Dwelling Older Adults J Holist Nurs, September 1, 2007; 25(3): 160 - 169. [Abstract] [PDF]
|
|
|
|
 |
|
 J. A. Burr, J. E. Mutchler, and F. G. Caro Productive Activity Clusters Among Middle-Aged and Older Adults: Intersecting Forms and Time Commitments J. Gerontol. B. Psychol. Sci. Soc. Sci., July 1, 2007; 62(4): S267 - S275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kim, J.-h. Kang, M.-A. Lee, and Y. Lee Volunteering Among Older People in Korea J. Gerontol. B. Psychol. Sci. Soc. Sci., January 1, 2007; 62(1): S69 - S73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Tang What Resources Are Needed for Volunteerism? A Life Course Perspective Journal of Applied Gerontology, November 1, 2006; 25(5): 375 - 390. [Abstract] [PDF]
|
|
|
|
|
|
M. A. Okun and J. Michel Sense of Community and Being a Volunteer Among the Young-Old Journal of Applied Gerontology, April 1, 2006; 25(2): 173 - 188. [Abstract] [PDF]
|
|
|
|
|
|
J. A. Burr, N. G. Choi, J. E. Mutchler, and F. G. Caro Caregiving and Volunteering: Are Private and Public Helping Behaviors Linked? J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2005; 60(5): S247 - S256. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
