Volunteer Retirement and Well-Being in Older Adults: A Longitudinal Study

Abstract

The extant literature on volunteering has focused primarily on the many benefits of volunteering for older adults. However, the question rarely investigated is whether these benefits dissipate when older adults retire from their volunteering. Given the U.S. policy context wherein volunteering is promoted as a solution to the problems of aging, this research investigates the association between the loss of one’s volunteering role through retirement and well-being. Utilizing three waves of the U.S.-based National Social Life, Health, and Aging Project (NSHAP) (2005–2016) and a fixed-effects modeling approach, we find that the well-being of older adults, measured as self-reported health, happiness, and depressive risk, is negatively associated with volunteer retirement. Our study contributes to the literature on well-being and volunteering for older adults and is the first study focusing on this critical transition point in the life of older volunteers. In addition, policymakers and organizations must broaden their focus to include not only the recruitment and retention of older adult volunteers but also the transition out of volunteering that many of them will eventually face.

Keywords

volunteering well-being older adults aging volunteer retirement

Introduction

Volunteering, a ubiquitous phenomenon in the United States, is defined as unpaid work done by an individual through an organization (Handy et al., 2000). There is considerable literature on volunteering and substantial evidence on the positive effects of volunteering (Konrath et al., 2012; Wilson, 2012). The extant literature at the intersection of gerontology and volunteering research has examined links between volunteering and well-being, finding that many of these benefits are particularly salient for older adults (e.g., Burr et al., 2011; Morrow-Howell et al., 2003; Musick et al., 1999; Piliavin & Siegl, 2015). These ideas align neatly with a narrative of aging productively, and U.S.-based researchers and policymakers alike have enthusiastically endorsed volunteering to stay active and healthy in older age (e.g., Hudson, 2007; National and Community Service Act of 1990, 2009).

Theorists have explained the health and well-being benefits of volunteering for older individuals using role theory (Biddle, 1986) and a life-course approach (Van Willigen, 2000). These approaches have suggested that volunteering is beneficial for older adults because it helps mitigate the loss of life roles that typically occur in later life, such as retirement from paid work, by acting as a role substitute (Russell et al., 2018). The loss and gain of different roles in life can be viewed as essential transition points in the lives of individuals, with positive or negative consequences for their identity, well-being, and health (Thoits, 2012, 2013). Another critical transition point is volunteer retirement, which occurs when older adults permanently discontinue their volunteer activities with an organization, thereby experiencing a loss of their volunteering role. Volunteer administrators have noted that in their experience, volunteer retirement is often followed by a decline in well-being (Russell et al., 2019). Although there is considerable literature suggesting how volunteers can be kept satisfied, retained, and motivated, volunteer retirement for older volunteers who age out of their volunteer roles is understudied (Shantz et al., 2014). These observations point to a logical, yet unstudied, research question: If volunteering boosts well-being in older adulthood, then is volunteer retirement accompanied by a subsequent decline in well-being?

Policymakers and health practitioners have continued to promote volunteering in the United States as a form of productive aging since the Serve America Act of 1990, which prioritized volunteer recruitment. However, they have shied away from addressing what happens when older volunteers retire from their volunteering. Do the benefits accrued through volunteering persist or decay following volunteer retirement? How should this transition be managed, and how should older volunteers be supported by organizations where they volunteer? These questions require consideration, as the demographic changes in the population suggest that Americans are projected to live longer by about 6 years, from 79.7 in 2017 to 85.6 in 2060 (Medina et al., 2020). This change in demographics suggests a need to pay greater attention to older volunteers’ life cycle as volunteers, not only when they commence volunteering but also when they retire from volunteering. Given these broader trends, we discuss our findings in the context of policy implications for organizations and their responsibilities to their older volunteers.

To investigate the relationship between well-being and volunteer retirement among older adults in the United States, we rely on longitudinal data from the National Social Life, Health, and Aging Project (NSHAP) (2005–2016). The NSHAP is a longitudinal, national study of health and social factors, aiming to understand the well-being of older Americans. We first examine whether the relationship between older adults’ volunteering and well-being is as expected for our dataset and analytic sample, given prevailing findings in the extant literature. We then turn to our investigation of volunteer retirement. To those ends, we address the following research questions:

Is volunteering among older adults associated with improvements in well-being over time?

Is retirement from volunteering associated with declines in older adults’ well-being over time?

Background and Significance

Volunteering and Well-Being in Older Adulthood

An essential topic of study in examining aging across the life course is the critical transitions that individuals experience as they age and how these transitions shape their identities, well-being, and health. As population projections mark 2030 as a demographic turning point for the United States, when one in every five Americans will be 65 years or older (Vespa et al., 2018), they bring into even greater significance the question of well-being in older adulthood and what it means to age well across the life course.

One of the many significant transitions that individuals undergo as they age is retirement from paid work. Researchers have examined the factors that influence individual well-being in their retirement (Wang & Shi, 2014). One factor that contributes to well-being is the individual’s choice of post-retirement activities. These activities may include further employment if there is a financial need, or their pursuit of unpaid pastimes, such as volunteering, leisure, and hobbies. Research has demonstrated a significant positive relationship between volunteering and well-being across the life course, with the most salient findings pertaining to older adults, who derive greater positive benefits from volunteering than younger adults (Tabassum et al., 2016; Van Willigen, 2000). Importantly, volunteering in older adulthood is neither universal nor uniform and varies across individuals and life stages with regard to activities undertaken, frequency and time committed, causes or types of organizations served, and other factors (Russell et al., 2020). Despite its variability, volunteering formally through organizations has been promoted by researchers and policymakers alike as a productive and beneficial activity in retirement (Gonzales et al., 2015; White House Conference on Aging, 2005, 2015).

Studies have investigated the link between volunteering and a variety of health and well-being outcomes, including mobility and physical activity (Pillemer et al., 2010); decreased mortality risk (Jenkinson et al., 2013; Musick et al., 1999; Oman et al., 1999); mental health, with an emphasis on depressive symptoms (Borgonovi, 2008; Li & Ferraro, 2006; Lum & Lightfoot, 2005; Musick & Wilson, 2003); physical health (Moen et al., 1992; Piliavin & Siegl, 2007); and other aspects of well-being, including life satisfaction (Binder, 2015; Binder & Freytag, 2013), positive affect, and purpose in life (Greenfield & Marks, 2004); and happiness (Borgonovi, 2008; Thoits & Hewitt, 2001). As such, the extant literature provides strong evidence for a positive relationship between volunteer activity and improved health and well-being among older adults, further demonstrated through meta-analyses by Jenkinson and colleagues (2013) and Okun et al. (2013).

Well-being is a multi-dimensional concept that is difficult to define (Veenhoven, 2000). Thus, scholars have noted the importance of examining more than one dimension of well-being to better capture these complexities (e.g., Baker et al., 2005), as evidenced by the breadth of investigated outcomes listed above. Happiness and self-rated health have been used to measure well-being in previous studies in the volunteering literature (e.g., Alfes et al., 2016; Borgonovi, 2008; Thoits, 2012, 2013). Boen and Yang (2016) also identify depressive risk to represent one aspect of psychosocial and emotional well-being, which has been linked to volunteering in previous studies (e.g., Musick & Wilson, 2003). According to a taxonomy created by Veenhoven (2000), these concepts fall into two different quadrants of well-being or quality of life: “life-ability” (health) and “appreciation of life” (depression and happiness) (p. 11). Thus, although related, these dimensions capture distinct aspects of well-being. As such, these findings in the literature point to the need to capture volunteers’ well-being using multiple measures.

Role Theory, Retirement, and Volunteering

Scholars have proposed various mechanisms that may link volunteering and well-being among older adults. Role theory has emerged as a critical explanation of why older adults who volunteer exhibit higher enhancements in well-being, both physical and mental, than not only their non-volunteer peers but also their younger volunteer counterparts (Greenfield & Marks, 2004). Role theory deals with major life transitions at all ages, including in and out of marriages, parenthood, and employment (e.g., Rotolo, 2000). Sociologists and social psychologists suggest that role theory explains why older adults who lose their roles and the concomitant social status as employees, parents, and caregivers find themselves at a loss within established social systems and their related norms and behaviors (Biddle, 1986; Van der Horst, 2016). For example, the transition from working life to retirement leads to role discontinuity or role ambiguity, which affects the well-being of individuals by causing stress (Biddle, 1986).

Volunteering can provide an opportunity for role enhancement that buffers against or overcomes the stress of both role discontinuity and role ambiguity (Cho et al., 2018). For example, Greenfield and Marks (2004) found empirical support for the role enhancement function of volunteering, which acts as a buffer between role loss and declines in well-being. Thus, a central application of role theory has been how volunteering provides role enhancement for older adults experiencing the role discontinuity and associated stressors of retirement from paid employment (Wang & Shultz, 2010). These benefits are markedly absent among younger adult volunteers, who do not experience the same role discontinuities as older adult volunteers and hence do not experience these buffering effects (Whillans et al., 2016).

The literature offers varied examples that illustrate the application of role theory. For instance, Moen and Fields (2002) find a positive association between volunteering and well-being among retired adults; notably, this association was significant only for those retired individuals not currently engaged in paid work, thereby suggesting a role enhancement function of volunteering. Sherman and Shavit (2012) further postulate that volunteering functions as a substitute for paid work, thus acting as the mechanism for maintaining well-being after retirement. Smith (2004) draws on role theory in a study of working adults’ perceptions of volunteering as a part of their future retirement, finding that they placed a high degree of role salience on volunteering in retirement. Finally, in a study of volunteering and well-being across the life course, Van Willigen (2000) proposes that the higher degree of discretion that older adults exercise when choosing to engage in volunteering contributes to the strength and significance of the link between volunteering and well-being observed among older adults.

These findings echo similar ones reported by Thoits (1992), who notes that discretionary or voluntary roles taken on by individuals are more likely to have positive impacts on psychological well-being than compulsory roles. While Thoits (1992) does not deal explicitly with volunteering, her findings nevertheless offer important insight into why the link between volunteering and well-being may be more salient than the link between other activities or roles, such as caregiving, and well-being. Later work by Thoits (2012, 2013) builds on these claims by investigating volunteering directly, suggesting that when individuals volunteer, they assume a beneficial identity for their well-being. Thoits argues that volunteering generates satisfaction and social rewards, which contribute to happiness, provides a sense of purpose and meaning, thereby alleviating depression, and creates opportunities for physical activity, thus positively affecting health. There is a congruence between the importance of the volunteer role for individuals and their self-identity.

In addition to these findings, the literature also suggests that the frequency of volunteering matters with regard to potential benefits to well-being (Thoits & Hewitt, 2001). For example, Thoits (2013) further identified a link between the frequency of volunteer activity and individuals’ well-being, finding that individuals who engage more frequently in a role will derive greater benefits from the role. She found that hours spent volunteering linked to greater volunteer identity salience, which acted as the mechanism through which volunteering related to better mental health. Musick and his colleagues (1999) found that moderate amounts of volunteering (measured in hours) were associated with lower mortality risks than non-volunteers. In addition, Alfes and her colleagues (2016) found that the intensity of volunteering related to happiness, as did Van Willigen (2000), who found that the positive effects of volunteering were related to the hours volunteered. Overall, there is strong evidence in the existing literature that as the frequency of volunteering increases, so too do the benefits to volunteers.

Scholars have also examined volunteering behaviors throughout the life cycle and noted that as individuals progress from adolescence, adulthood, middle age, and into older adulthood, their activities change, as does the probability and the intensity of their volunteering (Musick & Wilson, 2008). Nesbit (2012) identified that specific life events, such as the birth of a child, divorce, and widowhood, are also likely to affect volunteer activity. Although aging is not a specific life event, retirement from employment is, and past research suggests that there is an increased likelihood of older adults volunteering after retirement (Tang, 2016; Wilson, 2012). Another framework to consider in investigating how life events shape volunteer behavior is attachment theory, which suggests that individuals who feel disconnected or insecure due to a specific life event (e.g., unemployment) may be less likely to volunteer (Piatak, 2016). While attachment theory is underutilized in the volunteering literature (Wilson, 2012) and has not been used to examine the relationship between retirement from paid work and volunteering, this perspective may also help to illuminate the ways in which life events influence volunteering across the life course.

Moen (1996) writes, “The United States is a work-oriented society; paid work is the principal source of identity” (p. 131). Because of the centrality of work in the lives and identities of individuals (Svendsen, 2016), the impact of retirement and well-being continues to be an important area of study, especially in the United States. The promise of volunteering to mitigate the stress of the transition to retirement from the paid labor force, and the role discontinuity and ambiguity that this transition brings about, are encouraging in many ways. However, they engage with only one part of the story. In the next section, we introduce the concept of volunteer retirement to broaden this conversation.

Volunteer Retirement

Volunteer retirement is defined as the decision to withdraw from or discontinue volunteering (Russell et al., 2019). Volunteer retirement for older adults may reflect a series of complex and involuntary or voluntary events, such as failing health, lack of transportation to and from the volunteer site, or per the recommendation or encouragement of friends, family, and doctors. As is the case with retirement from paid work, individuals may decide to return to volunteering in some capacity in the future. However, the intent at the time of volunteer retirement is for the transition to be permanent, thereby distinguishing volunteer retirement from a leave of absence or a change to a different volunteer role.

In defining this concept, the term “retirement” is used purposefully as a means of highlighting the connection between volunteering in older adulthood and retirement from paid work. It both acknowledges the importance of work and productive activities in the lives and identities of individuals and connects this concept to the role theory perspective. If role theory points to a decline in well-being when individuals lose a salient role, then the loss of the volunteer role experienced through volunteer retirement may also be associated with stress and declines in well-being. Thus, volunteer retirement encompasses a role discontinuity.

To examine why older adults stop volunteering, Butrica et al. (2009) use longitudinal data that follow older individuals over 8 years to examine these transitions of older volunteers moving out of volunteering. They suggest that the increasing costs of volunteering may drive volunteers to quit. For example, changes in health or marital status may change the costs and benefits of volunteering to the individual, thereby driving their choices. Interestingly, the study found volunteering among older adults was mostly stable: Those who volunteered at baseline did so across the 8-year observation period, but among those that quit, factors such as residential moves, marriage, divorce, and widowhood were associated with quitting or breaks from volunteering. As individuals volunteer for longer periods of time, the odds that they quit decline, especially for those in good health.

Nevertheless, as individuals live longer and continue to volunteer across the life course, the likelihood that they eventually face a decision about whether to continue or to retire from their volunteering will increase. As it focuses mainly on active volunteers, the extant literature examining volunteering across the life cycle does not address the specific life event of volunteer retirement, and its impact on the well-being of older adults who retire from volunteering (e.g., Musick & Wilson, 2008; Nesbit, 2012; Tang, 2016). However, qualitative evidence from volunteer administrators suggests that older adults who volunteer regularly, especially for many years, experience the loss of this role in tangible ways (Russell et al., 2019). Thus, the dual issue of aging, followed by role discontinuity with respect to the volunteer role, suggests another transitional point, like the discontinuity of other roles worthy of further examination. In a person-driven profession like volunteer administration, understanding how volunteer retirement is associated with the health and well-being of individual volunteers can illuminate new considerations for the development of best practices. The present inquiry contributes to both the research literature and broader policy conversations about the relationships between volunteering and well-being.

Conceptual Framework

To connect the extant literature introduced above to our subsequent hypotheses and analysis, we introduce two conceptual frameworks. The literature on the benefits of volunteering for the well-being of older adults suggests the conceptual model outlined in Figure 1, in which volunteering provides role enhancement by substituting for roles lost in later life through retirement from paid work, widowhood, and others. According to Guiney and Machado (2018), there is potential bidirectionality in the relationship between volunteering and well-being, as illustrated in Figure 1. As such, we hypothesize that volunteer retirement could reverse the impact of the benefits of volunteering when individuals retire from their volunteer activities. Within the context of role theory, the transition out of the volunteering role represented by volunteer retirement would therefore lead to role discontinuity, as suggested by the conceptual model in Figure 2.

Figure 1.

Conceptual framework: Volunteering and well-being.

Figure 2.

Conceptual framework: Volunteer retirement and well-being.

We test the following hypotheses to investigate these relationships outlined in Figures 1 and 2 and suggested by the literature cited above. The first set of hypotheses pertains to our first research question on well-being and volunteering, and the second set pertains to our second research question on well-being and volunteer retirement in three time periods (T1–T3).

Hypotheses

Hypotheses for Question 1

Hypothesis 1A (H1A): Volunteering among older adults is positively associated with happiness, such that frequent engagement in volunteering is associated with higher levels of happiness over time (T1–T3) as compared to occasional or no engagement in volunteering.

Hypothesis 1B (H1B): Volunteering among older adults is positively associated with self-rated health, such that engagement in frequent volunteering is associated with higher levels of self-rated health over time (T1–T3) as compared to occasional or no engagement in volunteering.

Hypothesis 1C (H1C): Volunteering among older adults is negatively associated with depressive risk, such that engagement in frequent volunteering is associated with lower levels of depressive risk over time (T1–T3) as compared to occasional or no engagement in volunteering.

Hypotheses for Question 2

Hypothesis 2A (H2A): Volunteer retirement among older adults is associated with decreases in happiness over time (T1–T3).

Hypothesis 2B (H2B): Volunteer retirement among older adults is associated with decreased self-rated health over time (T1–T3).

Hypothesis 2C (H2C): Volunteer retirement among older adults is associated with increased depressive risk over time (T1–T3).

Research Design and Methods

Overview of the NSHAP Dataset

Data for this study come from the NSHAP (Waite, 2017; Waite et al., 2018a, 2018b). This study utilized all three waves of NSHAP data currently available (2005–2006; 2010–2011; 2015–2016). The NSHAP is one of few longitudinal, nationally representative datasets on older adults that systematically collects data on whether or not participants engaged in volunteering and the frequency of their volunteering. The original response rate for Wave 1 of the NSHAP survey was 75.5% (O’Muircheartaigh et al., 2009). This study includes individuals who participated in all three waves of the NSHAP study (n = 1,554) and excludes individuals who dropped out at either Wave 2 or Wave 3.

Outcome Variables: Dimensions of Well-Being

This study measured subjective well-being along three dimensions: happiness, self-rated health, and depressive risk. These measures have been used previously by researchers examining the relationship between volunteering and well-being (e.g., Alfes et al., 2016; Borgonovi, 2008; Musick & Wilson, 2003; Thoits, 2013). Happiness was measured in response to the question, “If you were to consider your life, in general, these days, how happy or unhappy would you say you are, on the whole?” Responses were measured on a scale of “unhappy usually” (1) to “extremely happy” (5). Participants rated both their mental health and their physical health from “poor” (1) to “excellent” (5). Self-rated health was the sum of scores for measures of self-rated mental health and self-rated physical health, generating a composite score for overall health, with scores ranging from 1 to 10. Depressive risk was measured using a shortened version of the Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977), comprising a series of questions related to the respondent’s thoughts and feelings in the past week. Example statements included “I felt depressed” and “I could not get going.” Respondents indicated how often they felt this way in the past week, from “rarely or none of the time” to “most of the time.” While the original scaling varied for these three dimensions of well-being, all variables were scaled uniformly from 0 to 1 for consistency and comparability (De Wit et al., 2015).

Predictor Variable: Formal Volunteering

The primary predictor is formal volunteering. Formal volunteering was captured in the NSHAP data set by asking participants about the frequency of their volunteer work in the past 12 months. It was operationalized in the analysis as both a categorical measure of the frequency of volunteering and a dichotomous measure at each time point. The measure of volunteer frequency was utilized to address the first research question and to test the first set of hypotheses. For this measure, the volunteering variable from the NSHAP was restructured into three categories: those who did not volunteer (vol = 0); those who volunteered occasionally, or less than once per month (vol = 1); and those who volunteered frequently or regularly, defined as volunteering at least once per month (vol = 2). These categories are in line with the volunteering literature (e.g., Meier & Stutzer, 2008) and reflect previous findings with regard to frequency of volunteering and well-being, as discussed above (Musick et al., 1999; Thoits, 2013; Van Willigen, 2000).

We constructed a third variable to address the second research question on “volunteer retirement.” Volunteer retirement is a dichotomous measure of volunteering that records when a person stops volunteering in either Wave 2 or Wave 3 of the dataset. This dichotomous measure was coded as 1 if the person stopped volunteering and 0 if the person continued volunteering in the subsequent wave. This coding allowed for greater ease of interpretation of model results following Boen and Yang (2016).

Control Variables: Time Variant and Time Invariant

Several individual characteristics have been found to influence volunteering, such as age, gender, race, ethnicity, and educational attainment (Musick & Wilson, 2008; Wilson, 2012), and these characteristics are typically included as control variables in analyses examining the consequences of volunteering. In the present study, only time-varying variables, such as age, were explicitly included in the model due to the use of fixed-effects regression modeling. In a fixed-effects model, time-invariant variables are “differenced away” and are thus excluded from model estimates (Allison, 2009).

The present analysis focuses on volunteering and age and their associations with well-being outcomes over time. Previous research has examined the relationship between well-being outcomes and age, with mixed results (Mroczek & Kolarz, 1998). For example, Frijters and Beatton (2012) found evidence for a U-shaped relationship between happiness and age. They also found that while happiness begins to increase around age 60, it then declines again after age 75. Because the youngest person in the present sample was 57 at T1, and the data span about 10 years, Frijters and Beatton’s (2012) findings offer a helpful comparison. Importantly, they find that the linear effect of age on happiness over time was consistently negative in their analyses of several large, population-based data sets (Frijters and Beatton, 2012, p. 529). Similarly, in their study of volunteering and well-being, Meier and Stutzer (2008) also found declining mean levels of well-being outcomes among participants in the German Socioeconomic Panel. Given previous evidence for a U-shaped relationship between age and specific well-being outcomes, we include age in years as both a linear and a quadratic term in the models.

Additional time-varying characteristics were also included as controls, including health status as measured by a cumulative total of pre-existing health conditions, whether or not the participant was married or living with a partner, and work status. Because we were interested in proxying the individual’s overall health status, rather than the effect of a specific health condition, health conditions were operationalized as the cumulative number of pre-existing conditions diagnosed by a physician that the participant reported at each time, following Boen and Yang (2016). We included seven different pre-existing conditions in this measure: arthritis, dementia, cancer, diabetes, stroke, hypertension, and heart disease. Because this number fluctuated over time, it was appropriate as a time-varying predictor in the analysis. Given the ways in which social integration may influence both health and well-being (Yang et al., 2015), we also included an additional control variable to capture individuals’ frequency of socializing with family and friends (1 = at least once per week). Marital status and work status were operationalized as dichotomous variables (1 = married or living with a partner; 1 = currently working).

Analytic Strategy

Data analysis for this study was conducted using Stata 17. All three waves of the dataset were cleaned and coded to include only those participants who responded to all three waves of the NSHAP survey (n = 1,554), representing approximately 52% of the original respondents at Wave 1. Reasons for attrition included death; refusal or nonresponse between waves; ineligibility for continued volunteering, such as moving to an assisted living facility; or illness (National Archive of Computerized Data on Aging, 2018). To determine whether there were significant differences between those who participated in all three waves and those who did not, the two groups were compared along demographic characteristics at Wave 1 (T1) using chi-square tests and t-tests (e.g., Baker et al., 2005; Boen & Yang, 2016; Yang et al., 2015). The results of this attrition analysis appear in the “Findings” section.

Regression models followed a fixed-effects approach that controlled for unobserved, time-invariant characteristics of individuals, helping to correct for omitted variable bias (Allison, 2009).¹ In addition, the fixed-effects approach is considered a more conservative modeling approach than random-effects models and produces less biased estimates than random effects; this is achieved by eliminating the between-person variation, which may be confounded by unobserved characteristics of the individuals (Allison, 2009).

Because the analysis included three waves of data, the analytic approach to fixed-effects modeling followed the mean deviation method, also called the conditional method (Allison, 2009). Listwise deletion was used to eliminate cases from the analysis where data were missing on the outcome variables. To retain the maximum amount of information, listwise deletion was carried out for each of the three models (i.e., happiness, self-rated health, and depressive risk) individually, resulting in a slightly different sample size for each model (reported in Tables 2 and 3).

To further illustrate how dimensions of well-being changed among the study sample over time, the grand mean scores for depressive risk, happiness, and self-rated health for each time point (T1–T3) were calculated. These mean scores were calculated for different subgroups of participants, according to their volunteering behaviors. These subgroups included those who never volunteered; those who always volunteered (“stable”); those who began the study as volunteers but withdrew at either Wave 2 or Wave 3 (“quitters”); and those who began the study as non-volunteers but began volunteering at either Wave 2 or Wave 3 (“joiners"). These categorizations of volunteer groups reflect strategies used by De Wit and colleagues (2015) and are helpful for the present study because they can visually represent the differences among participants over time, according to whether they volunteered, and if they withdrew from volunteering at different time points.

Findings

Descriptive Statistics

Table 1 presents the characteristics of the study sample at T1 (n = 1,554). Some of these variables, including gender and educational attainment, are known correlates of volunteering, are time-invariant, and thus not included in the fixed-effects regression analyses. However, they are included here to illustrate the characteristics of the study sample at T1.

Table 1.

Characteristics of the Study Sample (n = 1,554) at Wave 1.

Variable	M or %	SD	Min.	Max.
Outcomes
Happiness	0.701	0.190	0	1
Self-rated health	0.669	0.210	0	1
Depressive risk	0.163	0.164	0	1
Covariates
Age	66.7	6.81	57	85
Female	53.7%	—	0	1
Non-White	22.7%	—	0	1
Volunteer status	68.9%	—	0	1
Health conditions	1.74	1.25	0	7
Educational attainment
High school or equivalent	24.0%	—	0	1
Some college or vocational certification	30.8%	—	0	1
Bachelors or higher	26.7%	—	0	1
Married or living with a partner	69.2%	—	0	1
Work status	40.0%	—	0	1
Socialization with family and friends	55.4%	—	0	1

Attrition

The study sample at T1 represents the participants who remained in the study for all three waves and who were therefore eligible for inclusion in the analysis. This sample represents 51.7% of the original Wave 1 participants (n = 3,005). To examine the differences between the original participants and the individuals included in the final sample, we conducted both t-tests and chi-square tests. Findings indicated that individuals who dropped out before Wave 3 were significantly older (difference = 5.42 years, p < .001) and had more health conditions (difference = .475 health conditions, p < .001) at T1 than those who did not drop out.

In addition, volunteer status, gender, and educational attainment were not statistically independent from attrition (p < .05). Although these chi-square results were significant, they follow expectations. For example, women tend to live longer than men and therefore may be less likely to attrite over time. Similarly, socioeconomic status, such as higher levels of educational attainment, has been linked to better health outcomes as individuals age and longer life expectancies (Boen & Yang, 2016). Thus, those with higher educational attainment may be less likely to attrite. Finally, volunteering has not only been linked to improved health and reduced mortality risk as individuals age but also may exacerbate the “volunteer bias” already present among those who opt into research studies, such as the NSHAP (Salkind, 2010). As such, volunteers may be more likely to opt into study participation, and then to remain in subsequent waves of a study, than non-volunteers (Musick & Wilson, 2008).

These differences between the final analytic sample and those who dropped out at either Wave 2 or Wave 3 are not only in line with expectations, given prior research, but also a function of the study sample and the population of older adults. Nevertheless, they must be considered when interpreting the study findings, as they may introduce bias. For example, the analytic sample may be healthier, better educated, and more likely to have volunteered consistently across all three waves of the study than the general population, or, indeed, than the original NSHAP study sample at T1. Thus, the results of these analyses are expected to be more conservative estimates of the true relationships between changes in volunteering and changes in well-being outcomes over time than may be found in the general population of all older adults.

Question 1: Is volunteering among older adults associated with improvements in well-being over time?

Fixed-effects regression modeling was utilized to examine the relationship between volunteering and well-being over time. Volunteering was operationalized as a categorical variable, where occasional volunteers were distinguished from frequent volunteers and non-volunteers (reference category). The coefficients for each category illustrate the relationship between volunteering at different intensities on well-being, relative to the reference category of no volunteering. The analyses followed a stepwise approach, in which control variables were added to the model individually. Table 2 presents the results of the fully saturated models, with robust standard errors in parentheses.

Table 2.

Fixed-Effects Regression Results: Frequency of Volunteering.

Variable	Happiness	Self-rated health	Depressive risk
Number of observations	4,050	4,058	4,056
Independent variables
Volunteer frequency
Occasional	.014 (.010)	.022 (.008)**	−.002 (.007)
Frequent	.022 (.012)^†	.028 (.010)**	−.013 (.008)
Age	.009 (.007)	.024 (.006)***	−.022 (.005)***
Age-squared	−.00008 (.00005)^†	−.0002 (.00004)***	.0002 (.00003)***
Control variables
Health conditions	−.007 (.004)^†	−.011 (.003)***	−.0005 (.003)
Working	.019 (.010)^†	.022 (.008)**	−.010 (.007)
Married or living with partner	.069 (.018)***	−.006 (.012)	−.036 (.012)**
Socialization with family and friends	.030 (.008)***	−.004 (.007)	−.016 (.005)**
Intercept	.343	−.052	.952
Overall R²	.065	.064	.062

Note. Regression coefficients reflect rescaling of outcome variables from 0 to 1. Regression coefficients are unstandardized, and robust standard errors reported in parentheses.

†

p < .10. *p < .05. **p < .01. ***p < .001.

The findings varied across the three outcome variables. Those who volunteered frequently experienced marginally significant increases in happiness (p < .10) and significant increases in self-rated health (p < .01), relative to those who did not volunteer at all. We also found a significant positive relationship between self-rated health and volunteering over time for those who volunteered occasionally; however, we did not find a significant relationship between happiness and occasional volunteering. In addition, the relationship between volunteering any amount and depressive risk was not significant. The results show that, controlling for stable individual characteristics, frequent volunteering (at least once per month) is positively associated with self-rated health and happiness but not significantly associated with depressive risk, while occasional volunteering is positively associated with self-rated health but not significantly associated with happiness or depressive risk. Thus, H1A was partially supported, H1B was fully supported, and H1C was not supported.

We also tested the statistical difference between the coefficients for occasional and frequent volunteering across all three models using Wald tests. We found no significant differences between these coefficients and therefore failed to reject the equality hypothesis. This finding suggests that while there are significant differences between those who volunteer frequently and those who do not volunteer with regard to happiness and self-rated health, there are no significant differences between those who volunteered occasionally and those who volunteered frequently.

Question 2: Is retirement from volunteering associated with declines in older adults’ well-being over time?

We next examined the association between volunteer retirement and well-being outcomes over time. Figures 3 –5 illustrate changes in the grand means for each well-being outcome over time. Figures 3 and 4 show a general decline in happiness and self-rated health over time for all groups. In this way, they are aligned with previous findings of relationships between age and well-being outcomes (e.g., Frijters & Beatton, 2012).

Figure 3.

Happiness over time, by volunteering behavior.

Figure 4.

Self-rated health over time, by volunteering behavior.

Figure 5.

Depressive risk over time, by volunteering behavior.

In both Figures 3 and 4, those who volunteered in all three waves (“stable”) reported the highest mean levels of happiness and self-rated health over time, while those who never volunteered reported the lowest levels. The other two groups are both characterized by volunteering at least one time during the study. The “joiners” reported higher self-rated health over time than the “quitters,” as well as greater mean happiness in all three waves except Wave 2. However, it is essential to note that the “joiners” started higher on both dimensions of well-being at T1 as compared to the “quitters.” Moreover, the “quitters” experienced lower mean self-reported health at Wave 3 than those who had never volunteered in any of the waves. These observations suggest that older adults may leave their volunteering due to declining health and well-being, or due to various life circumstances. Thus, decisions to retire from volunteering may not only respond to changing feelings of well-being but also exacerbate or accelerate the decline. This possibility is explored through the regression analyses that follow.

Figure 5 illustrates the changing mean levels of depressive risk among the study sample over time.

As shown in Table 1, depressive risk deviated less over time than either happiness or self-rated health. Those who volunteered consistently at every time point reported the lowest levels of depressive risk, while those who never volunteered reported the highest levels on average. In both groups, the trends were also relatively flat, not changing much over time. By contrast, those who retired from volunteering (“quitters”) as a group exhibited more fluctuation over time, while those who joined volunteering at either Wave 2 or Wave 3 exhibited slightly higher and growing depressive risk over time, as compared to those who had volunteered in every wave. These findings are in line with expectations, namely, that those who volunteer consistently seem to experience greater benefits to their well-being over time. We explore this finding further in the analyses below.

To examine whether or not volunteer retirement was associated with participants’ well-being over time, we ran another series of fixed-effects models. The analyses followed the same stepwise approach as the previous set of analyses. In addition to examining the associations between age and volunteer retirement on well-being, the fully saturated models also controlled for pre-existing health conditions, work status, marital status, and frequency of socialization with family and friends.

As discussed above, because we are interested in whether or not volunteer retirement is negatively associated with well-being over time, volunteer retirement is coded as a binary variable, in which 0 indicates volunteering and 1 indicates volunteer retirement. Table 3 presents the results for the fully saturated models; robust standard errors appear in parentheses.

Table 3.

Fixed-Effects Regression Results: Volunteer Retirement.

Variable	Happiness	Self-rated health	Depressive risk
Number of observations	4,050	4,058	4,056
Independent variables
Volunteer retirement	−.015 (.009)^†	−.024 (.008)**	.006 (.007)
Age	.009 (.007)	.024 (.006)***	−.022 (.005)***
Age-squared	−.00008 (.00005)^†	−.0002 (.00004)***	.0002 (.00003)***
Control variables
Health conditions	−.007 (.004)^†	−.011 (.003)***	−.0003 (.003)
Working	.019 (.010)^†	.022 (.008)**	−.009 (.007)
Married or living with partner	.069 (.018)***	−.005 (.012)	−.036 (.012)**
Socialization with family and friends	.030 (.008)***	−.003 (.006)	−.017 (.005)***
Intercept	.350	−.037	.962
Overall R²	.063	.061	.057

Note. Regression coefficients reflect rescaling of outcome variables from 0 to 1. Regression coefficients are unstandardized, and robust standard errors reported in parentheses.

†

p < .10. *p < .05. **p < .01. ***p < .001.

In line with expectations, the findings suggest that volunteer retirement was negatively associated with both happiness and self-rated health over time, such that volunteer retirement was related to declines in both happiness and self-rated health. As expected, these models demonstrated a negative and significant linear relationship between age and both happiness and self-rated health. Thus, H2A and H2B were supported.

Contrary to expectations, volunteer retirement was not significantly related to depressive risk over time. While the coefficient for volunteer retirement was in the expected direction (i.e., volunteer retirement and depressive risk are positively associated), it was not significant in the fully saturated model, when all time-variant control variables had been added. While age and other measures of social interaction (married or living with a partner and frequency of socialization with family and friends) were significantly and negatively associated with depressive risk, volunteer retirement was not. Thus, H2C was not supported.

Subgroup Analyses

Results from the fixed-effects models imply that entering and exiting volunteering is symmetrical, meaning that if an individual begins or stops volunteering, the respective increase or decrease associated with subjective well-being will be equal in magnitude. The results presented in Table 3 do not account for differences in patterns of volunteering over time. Instead, the coefficient for volunteer retirement in the fixed-effects models indicates the average deviation in subjective well-being for the individual around a person-specific mean, controlling for both observed, time-varying predictors and unobserved, time-invariant characteristics. However, as illustrated in Figures 3 –5, the relationships between volunteering and subjective well-being outcomes appear to differ, to varying extents, among different subgroups of participants.

To determine whether there were differences among subgroups of participants, we conducted a series of sensitivity analyses. To create the subgroups, participants were differentiated according to their patterns of volunteering over the three waves in the same manner as before: those who never volunteered (“Never Volunteered”); those who always volunteered (“Stable”); those who started out as non-volunteers but joined at either T2 or T3 (“Joiners”); and those who started out as volunteers but withdrew at either T2 or T3 (“Quitters”). Because this study is concerned with how changes in volunteering are associated with subjective well-being over time, and because fixed-effects regression does not account for those whose volunteer status remained constant (i.e., time-invariant) over time, these sensitivity analyses were conducted only for the “Joiners” and the “Quitters.” To do this, we used fixed-effects models with robust standard errors. All models were fully saturated, controlling for age, pre-existing health conditions, work status, marital status, and frequency of socialization with family and friends. Table 4 summarizes the results of these analyses by providing the coefficients for volunteer retirement for each subjective well-being outcome variable by subgroup. The coefficients indicate the relationship between volunteer retirement and subjective well-being outcomes for each of the subgroups.

Table 4.

Fixed-Effects Regression Results for Subgroup Analyses, Fully Saturated Models.

Variable	Subgroup 1: Joiners	Subgroup 2: Quitters
Number of observations	479	738
Subjective well-being outcome
Happiness	−.003 (.019)	−.042 (.018)*
Self-rated health	−.011 (.013)	−.026 (.012)*
Depressive risk	−.015 (.013)	.038 (.014)**

Note. Regression coefficients reflect rescaling of outcome variables from 0 to 1. Regression coefficients are unstandardized, and robust standard errors reported in parentheses.

†

p < .10. *p < .05. **p < .01. ***p < .001.

The results presented in Table 4 suggest that the relationships between volunteering and subjective well-being over time are different between these two subgroups of participants. For the “Joiners,” the relationships between subjective well-being outcomes and volunteer retirement over time were insignificant. These findings indicate that among those who started out as non-volunteers but joined in at either T2 or T3, changes in volunteer behavior over time were not significant predictors of changes in subjective well-being for happiness, self-rated health, or depressive risk.

By contrast, for the “Quitters,” the relationships between well-being outcomes and volunteer retirement over time were significant for happiness, self-rated health, and depressive risk. These findings indicate that among “Quitters” who volunteered in T1 but retired at either T2 or T3, volunteer retirement was negatively related to both happiness and self-rated health and positively related to depressive risk. Overall, the results presented in Table 4 suggest that the relationships between volunteering and well-being outcomes are not symmetrical and are dependent on older adults’ previous volunteering behaviors.

Although the two subgroups differed in size, with the group of “Quitters” (n = 246) outnumbering the “Joiners” (n = 160), the use of fixed-effects modeling helps to control for the effect of any time-invariant differences between the two groups, which might have otherwise biased any comparisons. Likewise, the inclusion of age, health conditions, work status, marital status, and frequency of socialization with family and friends as time-varying predictors in the models also controls for these characteristics of the participants, strengthens inferences that can be made, and enables comparisons between the two subgroups. Finally, the inclusion of the number of pre-existing health conditions diagnosed by a physician to control for time-varying health status (e.g., Boen & Yang, 2016) helps to better parse out the relationship between volunteering and subjective well-being over time.

Discussion and Future Directions

The literature on volunteering and well-being consistently finds evidence for the benefits of volunteering, especially among older adults. Drawing on these findings and the notion of productive aging, researchers and policymakers have promoted volunteering in later life as a means to address the problem of aging. Such promotion has led to the creation of federal policies and programs geared toward promoting volunteering in older adulthood (Gonzales et al., 2015), as well as an abundance of literature on recruitment and retention of older adult volunteers at the organizational level (Russell et al., 2019). While this study acknowledges the importance of this existing evidence on the benefits of volunteering in older adulthood, it also broadens our consideration of the experiences of older adult volunteers by examining the phenomenon of volunteer retirement and its association with well-being.

The findings concerning our first research question show that volunteering is positively related to happiness and self-rated health over time, overall replicating findings in the extant literature regarding the relationship between well-being and volunteering (e.g., Jenkinson et al., 2013; Okun et al., 2013; Piliavin & Siegl, 2015). However, the findings were not the same for happiness and self-rated health, with more compelling evidence found linking self-rated health and volunteering over time. Notably, we found significant positive relationships between self-rated health and both frequent and occasional volunteering but only found a significant positive relationship between happiness and frequent volunteering. These findings suggest that engaging in regularly scheduled rather than occasional volunteering will have the most significant influence on various dimensions of well-being relative to no volunteering (Alfes et al., 2016; Musick et al., 1999; Thoits, 2013; Van Willigen, 2000).

However, these findings pertain to the comparison between frequent or occasional volunteering and the reference category of no volunteering. Using Wald tests, we also compared the categories of frequent and occasional volunteering to each other. The results indicated that there were no statistically significant differences between these categories, suggesting that continuing (or beginning) to volunteer over time is more important than increasing the frequency of volunteering among older individuals. Given our conceptualization of volunteering as a possible role substitute in later life, we find these results consistent with the notion that the stable presence of the volunteer role in individuals’ lives contributes more to well-being as opposed to the number of hours or occurrences.

In examining the second research question, this study also investigated the novel concept of volunteer retirement and therefore contributes to the literature on volunteering and well-being in older adulthood both conceptually and empirically. It provides empirical evidence for the first time indicating that volunteer retirement is associated negatively with well-being. Although previous studies have examined individuals’ transitions in and out of volunteering, their reasons for doing so, and the factors that may contribute to such transitions (Butrica et al., 2009; Musick & Wilson, 2008; Nesbit, 2012), these studies have not examined the association of such transitions with well-being. Our results build on these previous studies and point to the need to give serious consideration to volunteer retirement in the volunteer life cycle of older adults, alongside recruitment and retention, as a key transition warranting further study.

Overall, this study adds to the literature on the complex relationship between volunteering and well-being among older adults over time. The results of our analyses demonstrated that changes in volunteer status were significantly related to the well-being of older adults in the NSHAP study, even when controlling for time-invariant characteristics, as well as time-varying predictors commonly associated with volunteering. Because fixed-effects regression modeling produces more conservative effect sizes (coefficients) than random-effects regression, the significant findings concerning happiness and self-rated health may be smaller in magnitude but represent more robust representations of the true relationship between volunteer retirement and well-being over time.

Moreover, the inclusion of pre-existing health conditions as time-varying controls in these models suggests that the relationships between volunteer retirement and well-being outcomes occur over and above changes in health status. Undoubtedly, the relationship between health, well-being, and volunteering among older adults is a complex one. Changes in volunteering behavior may both reflect a response to changing health and well-being and, perhaps, contribute to such changes over time among older adults. Given the possibility that volunteering functions as a role substitute for other social roles, as suggested by role theory, this assertion is a plausible one. Volunteering can become a crucial part of the lives of older adults and, thus, not something that individuals will give up readily.

Our findings are also congruent with our proposed conceptual frameworks (Figures 1 and 2). Drawing on role theory, these frameworks illustrate an association of the benefits of the role enhancement function of volunteering, indicating, as other researchers have found (e.g., Thoits, 2012, 2013), that for older adults, there is a significant and positive association between volunteering and well-being. We find this significant positive association between volunteering and both happiness and self-reported health (supporting H1A and H1B). However, although the direction of the relationship is in line with the hypothesized direction, we do not find a significant relationship between volunteering and depressive risk over time (H1C). The relative stability of depressive risk over time in this sample (see Table 1 and Figure 4) may have contributed to this finding. Due to the longitudinal nature of the study, attrition may also explain this result. Attrition contributes to more conservative results, in that those who experience extremes at the low end of the outcome variables were more likely to drop out over time than those who experienced higher levels. Thus, this bias in the data could contribute to the lack of significance concerning depressive risk and volunteering over time.

Our second conceptual model suggests that the effects of the role enhancement function of volunteering dissipate when individuals retire from volunteering due to role loss. Volunteer retirement is a specific life event that leads to role discontinuity associated with stressors (Wang & Shultz, 2010). As expected, our hypotheses are supported for happiness and self-reported health, showing significant decreases (H2A and H2B) but not supported for depressive risk (H2C). As stated above, the relative stability of depressive risk over time and potential sources of bias in the dataset could have contributed to this result.

Overall, our findings support the application of role theory as one framework through which to explain the link between volunteering and well-being and strengthen the role enhancement model for volunteering. Previous studies have posited that volunteering acts as a substitute for other roles that individuals typically lose in later life, emphasizing the importance of volunteering for buffering against the negative consequences of role loss associated with retirement, widowhood, and others (Russell et al., 2018). These studies emphasize beginning and continuing to volunteer and their implications for older adults’ health and well-being. In broadening the discussion of volunteering in older adulthood to consider not only beginning and continuing to volunteer but also volunteer retirement, we expand the application of role theory to suggest that in the process of substituting for other roles, volunteering inevitably takes on a greater degree of salience in the lives of older adults. As such, the loss of the volunteering role, just like the loss of other roles, becomes another critical transition in older adulthood. Thus, in applying role theory to the study of volunteering in older adulthood, it is important to consider volunteering not only as a substitute for other roles but also as a primary role in and of itself, with significant implications for the individual’s identity and well-being (Russell et al., 2022).

In addition to their contributions to the existing literature on volunteering and well-being, the findings also suggest policy and practice implications. As stated above, shifting demographics in the United States have given rise to a number of socioeconomic concerns related to population aging (Medina et al., 2020). Encouraged by research conducted by both volunteering scholars and gerontologists, policymakers have consistently promoted volunteering as a viable policy and public health intervention to promote healthy aging (e.g., AmeriCorps, 2019; National and Community Service Act of 1990, 2009; White House Conference on Aging, 2005, 2015). By exhorting older adults to volunteer, however, policymakers cannot ignore them as they encounter the transition of volunteer retirement. Both policymakers and organizations must broaden their focus to include not only the recruitment and retention of older adult volunteers but also the transition out of volunteering that many of them will eventually face. Yet, policy documents and federal programs supporting elder volunteering are notably silent on this issue, and the literature on volunteer management overwhelmingly continues to emphasize recruitment and retention strategies, with little to no consideration of volunteer transitions at any age.

Given practitioner perspectives that volunteer retirement is accompanied by individuals’ decline (Russell et al., 2019), and given the findings from the present analyses, volunteer administrators and other staff will need to consider their role in helping to manage this transition. Organizations that recruit older volunteers should consider adopting new policies and procedures to help support their staff in these efforts. Such policies should be viewed as part of a broader strategic plan concerning the role and management of volunteers within the organization. Moreover, they may be vital for organizations without a dedicated volunteer administrator, where the task of managing volunteers is spread out among multiple staff or even other volunteers, as these situations may precipitate more significant concerns about organizational liability, risk, and human resources management.

Study Limitations

This study, like all studies, has some limitations, which in turn suggest possibilities for future research. First, many scholars have pointed out the difficulty of parsing out the true associations between volunteering and well-being, thereby cautioning against overstating the magnitude or causal direction of the relationship (e.g., De Wit et al., 2015). Fixed-effects models are considered more robust approaches for dealing with endogeneity bias than random effects (Halaby, 2004; Milner et al., 2016), but they do not eliminate the issue. Endogeneity bias thus remains a concern in the present analysis, in which the use of certain techniques to ameliorate it, such as auxiliary instrumental variables, were not possible due to the lack of suitable variables in the NSHAP dataset (Bollen, 2012). Given these constraints, which are common with observational datasets, we take care not to overstate our conclusions, as there may be unmeasured or unobserved variables that are not captured in our data. Nevertheless, we view the findings as sufficiently intriguing to justify further research to identify the mechanism for causal directions. Thus, we recommend further study of volunteer retirement and its role as a critical transition in the lives of older adults, using data that comprehensively measure the impact on well-being.

Second, the quantitative study of volunteering, a social phenomenon, often represents trade-offs for researchers interested in examining nuances of the individual volunteer experience and capturing its effects over time. In the present research, the analyses were limited in their scope by the availability of detailed longitudinal data on volunteering among older adults. Most data sets that ask individuals in-depth questions about their volunteering, such as the types of tasks performed or their reasons for beginning or ending their involvement, are cross-sectional and cannot be used to investigate the transitions that occur over time in an individuals’ volunteer behaviors. While this study gained the strengths of longitudinal, nationally representative panel data, it was limited by a lack of specificity about individuals’ volunteering. The reasons for volunteer retirement may be particularly important for the well-being of older adults (e.g., Butrica et al., 2009), and the inability to include them in the present analysis is undoubtedly a limitation. For instance, whether or not individuals retire from volunteering due to circumstances that they feel are within their control (e.g., wanting to spend more time on other activities) versus circumstances that they perceive as beyond their control (e.g., health concerns, coupled with pressure from family, friends, doctors, or the organization itself), or whether this ultimately represents a permanent transition or one they decide to reverse, could play a critical role in determining the extent to which volunteer retirement is associated with declines in their well-being (Russell et al., 2019). Another factor to consider is whether or not the decision to retire from volunteering ultimately represents a permanent transition for older adults or if they look to return to the same or a substitute volunteer role at a later point in time. Future studies might attempt to address these limitations by utilizing a quasi-experimental design to examine the influence of reasons for volunteer retirement on older adults’ well-being or a time series analysis to allow for a more nuanced look at volunteer activities across shorter units of time.

Finally, the study data are collected from a sample of American older adults. Our findings and conclusions are therefore limited to the U.S. context. Future research should seek to expand the investigation of volunteer retirement and its implications for older volunteers and organizations and volunteer administrators, to other countries, where both the broader policy context and the meanings and experiences of volunteering and retirement may differ.

Future Directions for Volunteer Management

This study raises several important considerations for volunteer management. Previous research and practice have focused largely on the recruitment and retention of older adult volunteers. The findings from this study push us to consider how organizations should manage older adult volunteers at all stages of their volunteering, including during transitions out of volunteering. Is it the responsibility of volunteer administrators and other nonprofit- and public-sector staff to care for the well-being of volunteers who may no longer be able to contribute “productively” to their organizations? If so, what organizational policies and professional practices should be developed to assist with this challenge? In exploring the concept of volunteer retirement and its relationship to well-being, this study has sought to broaden the conversation around older adult volunteering and to encourage further consideration of these critical managerial questions. Future research should focus on developing and evaluating best practices to support both older adult volunteers and volunteer managers in planning for and navigating volunteer retirement. Such efforts will contribute to improve not only volunteer experiences for older adults but also stronger volunteer engagement strategies for staff and organizations.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Allison R. Russell

Femida Handy

Notes

Author Biographies

Allison R. Russell is an assistant professor in the Nonprofit and Public Management Program at the School of Economic, Political, and Policy Sciences, University of Texas at Dallas. Her research and teaching focus on nonprofit management, volunteering, ethics, and social impact.

Courtney E. Boen is an assistant professor of sociology at the University of Pennsylvania. Her research focuses primarily on the social determinants of population health inequality, with particular attention to the social factors producing racial and socioeconomic health inequalities.

Femida Handy is a professor at the School of Social Policy and Practice at the University of Pennsylvania. An economist by training, her work encompasses the economics of the nonprofit sector, volunteering, philanthropy, nonprofit management, environmental issues entrepreneurship, and microfinance.

References

Alfes

Shantz

Bailey

(2016). Enhancing volunteer engagement to achieve desirable outcomes: What can nonprofit employers do? VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 27(2), 595–617.

Allison

P. D.

(2009). Fixed effects regression methods in SAS. https://pdfs.semanticscholar.org/84f5/55569662b8c4882b213cd13f75622eaf495e.pdf

AmeriCorps. (2019, March 5). Federal agency awards $13.6 million to support senior volunteer programs in 150 communities. https://www.nationalservice.gov/newsroom/press-releases/2019/federal-agency-awards-136-million-support-senior-volunteer-programs-150

Baker

L. A.

Cahalin

L. P.

Gerst

Burr

J. A.

(2005). Productive activities and subjective well-being among older adults: The influence of number of activities and time commitment. Social Indicators Research, 73, 431–458.

Biddle

B. J.

(1986). Recent developments in role theory. Annual Review of Sociology, 12, 67–92.

Binder

Freytag

(2013). Volunteering, subjective well-being and public policy. Journal of Economic Psychology, 34, 97–119.

Binder

(2015). Volunteering and life satisfaction: A closer look at the hypothesis that volunteering more strongly benefits the unhappy. Applied Economic Letters, 22(11), 874–885.

Boen

Yang

Y. C.

(2016). The physiological impacts of wealth shocks in late life: Evidence from the Great Recession. Social Science & Medicine, 150, 221–230.

Bollen

K.A.

(2012). Instrumental variables in sociology and the social sciences. Annual Review of Sociology, 38, 37–72.

10.

Borgonovi

(2008). Doing well by doing good: The relationship between formal volunteering and self-reported health and happiness. Social Science & Medicine, 66, 2321–2334.

11.

Burr

J. A.

Tavares

Mutchler

J. E.

(2011). Volunteering and hypertension risk in later life. Journal of Aging and Health, 23(1), 24–51.

12.

Butrica

B.A.

Johnson

R.W.

Zedlewski

S.R.

(2009). Volunteer dynamics among older Americans. The Journals of Gerontology Series B Psychological Sciences and Social Sciences, 64(5), 644–655.

13.

Cho

Kim

Park

Jang

(2018). Post-retirement work and volunteering by poverty groups informed by role theory. Journal of Gerontological Social Work, 61(3), 243–260.

14.

De Wit

Bekkers

Karamat Ali

Verkaik

. (2015). Welfare impacts of participation. ITSSOIN. European Commission, DG Research.

15.

Frijters

Beatton

(2012). The mystery of the U-shaped relationship between happiness and age. Journal of Economic Behavior and Organization, 82, 525–542.

16.

Gonzales

Matz-Costa

Morrow-Howell

(2015). Increasing opportunities for the productive engagement of older adults: A response to population aging. The Gerontologist, 55(2), 252–261.

17.

Greenfield

E. A.

Marks

N. F.

(2004). Formal volunteering as a protective factor for older adults’ psychological well-being. Journal of Gerontology: Social Sciences, 59B(5), S258–S264.

18.

Guiney

Machado

(2018). Volunteering in the community: Potential benefits for cognitive aging. The Journals of Gerontology: Series B, 73(3), 399–408.

19.

Halaby

C. N.

(2004). Panel models in sociological research: Theory into practice. Annual Review of Sociology, 30, 507–544.

20.

Handy

Cnaan

R. A.

Brudney

Meijs

Ascoli

Ranade

(2000). Public perception of “who is a volunteer”: An examination of the net-cost approach from a cross-cultural perspective. Voluntas: International Journal of Voluntary & Nonprofit Organizations, 11(1), 45–65.

21.

Hudson

R. B.

(2007). Aging in a public space: The roles and functions of civic engagement. Generations, 30(4), 51–58.

22.

Jenkinson

C. E.

Dickens

A. P.

Jones

Thompson-Coon

Taylor

R. S.

Rogers

Bambra

C. L.

Lang

Richards

S. H.

(2013). Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers. BMC Public Health, 13, 773.

23.

Konrath

Fuhrel-Forbis

Lou

Brown

(2012). Motives for volunteering are associated with mortality risk in older adults. Health Psychology, 31(1), 87–96

24.

Ferraro

K. F.

(2006). Volunteering in middle and later life: Is health a benefit, barrier, or both? Social Forces, 85(1), 497–519.

25.

Lum

T. Y.

Lightfoot

(2005). The effects of volunteering on the physical and mental health of older people. Research on Aging, 27(1), 31–55.

26.

Medina

Sabo

Vespa

(2020). Living longer: Historical and projected life expectancy in the United States, 1960 to 2060. US Department of Commerce, US Census Bureau.

27.

Meier

Stutzer

(2008). Is volunteering rewarding in itself? Economica, 75, 39–59.

28.

Milner

Aitken

Kavanagh

LaMontagne

A. D.

Petrie

(2016). Persistent and contemporaneous effects of job stressors on mental health. Occupational and Environmental Medicine, 73, 787–793.

29.

Moen

(1996). A life course perspective on retirement, gender, and well-being. Journal of Occupational Health Psychology, 1(2), 131–144.

30.

Moen

Dempster-McClain

Williams

R.M.

(1992). Successful aging: A life-course perspective on women’s multiple roles and health. American Journal of Sociology, 97(6). https://doi.org/10.1086/229941

31.

Moen

Fields

(2002). Midcourse in the United States: Does unpaid community participation replace paid work? Ageing International, 27(3), 21–48.

32.

Morrow-Howell

Hinterlong

Rozario

P. A.

Tang

(2003). Effects of volunteering on the well-being of older adults. Journal of Gerontology: Social Sciences, B, 58(3), S137–S145.

33.

Mroczek

D. K.

Kolarz

C. M.

(1998). The effect of age on positive and negative affect: A developmental perspective on happiness. Personality and Social Psychology, 75(5), 1333–1349.

34.

Musick

M. A.

Herzog

A. R.

House

J. S.

(1999). Volunteering and mortality among older adults: Findings from a national sample. Journal of Gerontology: Social Sciences, B, 54(3), S173–S180.

35.

Musick

M. A.

Wilson

(2003). Volunteering and depression: The role of psychological and social resources in different age groups. Social Science & Medicine, 56, 259–269.

36.

Musick

M. A.

Wilson

(2008). Volunteers: A Social Profile. Indiana University Press.

37.

National and Community Service Act of 1990. (2009, May 19). https://www.nationalservice.gov/sites/default/files/documents/1990_serviceact_as%20amended%20through%20pl%20111-13.pdf

38.

National Archive of Computerized Data on Aging. (2018, February 13). National Social Life, Health, and Aging Project (NSHAP): Wave 2 and Partner Data Collection (ICPSR 34921). https://www.icpsr.umich.edu/icpsrweb/DSDR/studies/34921

39.

Nesbit

(2012). The influence of major life cycle events on volunteering. Nonprofit and Voluntary Sector Quarterly, 41(6), 1153–1174.

40.

Okun

M. A.

Yeung

E. W.

Brown

(2013). Volunteering by older adults and risk of mortality: A meta-analysis. Psychology and Aging, 28(2), 564–577.

41.

Oman

Thoresen

C. E.

McMahon

(1999). Volunteering and well-being among the community-dwelling elderly. Journal of Health Psychology, 4(3), 301–316.

42.

O’Muircheartaigh

Eckman

Smith

(2009). Statistical design and estimation for the national social life, health, and aging project. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 64(Suppl1), i12–i19. https://doi.org/10.1093/geronb/gbp045

43.

Piatak

J. S.

(2016). Time is on my side: A framework to examine when unemployed individuals volunteer. Nonprofit and Voluntary Sector Quarterly, 45(6), 1169–1190.

44.

Piliavin

J.A.

Siegl

(2007). Health benefits of volunteering in the Wisconsin Longitudinal Study. Journal of Health and Social Behavior, 48, 450–464.

45.

Piliavin

J. A.

Siegl

(2015). Health and well-being consequences of formal volunteering. In Schroeder

D. A.

Graziano

W.G.

(Eds.), The Oxford handbook of prosocial behavior. https://doi.org/10.1093/oxfordhb/9780195399813.013.024

46.

Pillemer

Fuller-Rowell

T. E.

Reid

M. C.

Wells

N. M.

(2010). Environmental volunteering and health outcomes over a 20-year period. The Gerontologist, 50(5), 594–602.

47.

Radloff

L. S.

(1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurements, 1, 385–401.

48.

Rotolo

(2000). A time to join, a time to quit: The influence of life cycle transitions on voluntary association membership. Social Forces, 78(3), 1133–1161.

49.

Russell

A. R.

Heinlein Storti

M. A.

Handy

(2019). Managing volunteer retirement among older adults: Perspectives of volunteer administrators. Journal of Public and Nonprofit Affairs, 5(1), 95.

50.

Russell

A. R.

Kim

Handy

Gellis

(2020). Formal versus informal volunteering and wellbeing: Does volunteering type matter for older adults? Voluntary Sector Review, 11(3), 317–336.

51.

Russell

A. R.

Nyame-Mensah

De Wit

Handy

(2018). Volunteering and wellbeing among ageing adults: A longitudinal analysis. Voluntas. https://doi.org/10.1007/s11266-018-0041-8

52.

Russell

A. R.

Storti

M. H.

Handy

(2022). Volunteering Retirement and well-being: Evidence from older adult volunteers. International Journal of Community Wellbeing, 5(2), 475–495.

53.

Salkind

N. J.

(2010). Volunteer bias. Encyclopedia of Research Design. https://dx-doi-org-s.web.bisu.edu.cn/10.4135/9781412961288.n492

54.

Shantz

Saksida

Alfes

(2014). Dedicating time to volunteering: Values, engagement, and commitment to beneficiaries. Applied Psychology: An International Review, 63(4), 671–697.

55.

Sherman

Shavit

(2012). How the life cycle hypothesis explains volunteering during retirement. Ageing and Society, 32(8), 1360–1381.

56.

Smith

D. B.

(2004). Volunteering in retirement: Perceptions of midlife workers. Nonprofit and Voluntary Sector Quarterly, 33(1), 55–73.

57.

Svendsen

(2016). Work (2nd edn.). Routledge.

58.

Tabassum

Mohan

Smith

(2016). Association of volunteering with mental well-being: A lifecourse analysis of a national population-based longitudinal study in the UK. BMJ Open, 6. https://doi.org/10.1136/bmjopen-2016-011327

59.

Tang

(2016). Retirement patterns and their relationship to volunteering. Nonprofit and Voluntary Sector Quarterly, 45(5), 910–930.

60.

Thoits

P. A.

(1992). Identity structures and psychological well-being: Gender and marital status comparisons. Social Psychology Quarterly, 55(3), 236–256.

61.

Thoits

P. A.

(2012). Role-identity salience, purpose and meaning in life, and well-being among volunteers. Social Psychology Quarterly, 75(4), 360–384.

62.

Thoits

P. A.

(2013). Volunteer identity salience, role enactment, and well-being: Comparisons of three salience constructs. Social Psychology Quarterly, 76(4), 373–398.

63.

Thoits

P.A.

Hewitt

L.N.

(2001). Volunteer work and well-being. Journal of Health and Social Behavior, 42(2), 115–131.

64.

Van der Horst

. (2016, July 27). Role theory. Oxford Bibliographies. https://doi.org/10.1093/OBO/9780199756384-0175

65.

Van Willigen

. (2000). Differential benefits of volunteering across the life course. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 55(5), S308–318.

66.

Veenhoven

(2000). The four qualities of life: Ordering concepts and measures of the Good Life. Journal of Happiness Studies, 1, 1–39.

67.

Vespa

Armstrong

D. M.

Medina

(2018). Demographic turning points for the United States: Population projections for 2020 to 2060. US Department of Commerce, Economics and Statistics Administration, US Census Bureau.

68.

Waite

(2017, November 25). National Social Life, Health and Aging Project (NSHAP): Wave 3. Inter-university Consortium for Political and Social Research. https://doi.org/10.3886/ICPSR36873.v1

69.

Waite

L. J.

Cagney

K. A.

Dale

Huang

Laumann

E. O.

McClintock

M. K.

O’Muircheartaigh

C. A.

Schumm

L. P.

Cornwell

(2018a, February 13). National Social Life, Health, and Aging Project (NSHAP): Wave 2 and Partner Data Collection. Inter-university Consortium for Political and Social Research. https://doi.org/10.3886/ICPSR34921.v2

70.

Waite

L. J.

Laumann

E. O.

Levinson

W. S.

Lindau

S. T.

O’Muircheartaigh

C. A.

(2018b, January 29). National Social Life, Health, and Aging Project (NSHAP): Wave 1. Inter-university Consortium for Political and Social Research. https://doi.org/10.3886/ICPSR20541.v7

71.

Wang

Shi

(2014). Psychological research on retirement. Annual Review of Psychology, 65, 209–233.

72.

Wang

Shultz

K. S.

(2010). Employee retirement: A review and recommendations for future investigation. Journal of Management, 36(1), 172–206.

73.

Whillans

A. V.

Seider

S. C.

Chen

Dwyer

R. J.

Novick

Gramigna

K. J.

. . . Dunn

E. W.

(2016). Does volunteering improve well-being? Comprehensive Results in Social Psychology, 1(1-3), 35–50.

74.

White House Conference on Aging. (2005). Final Report. http://nicoa.org/wp-content/uploads/2014/05/2005-WHCOA-Final-Report.pdf

75.

White House Conference on Aging. (2015). Final Report. https://whitehouseconferenceonaging.gov/2015-WHCOA-Final-Report.pdf

76.

Wilson

(2012). Volunteerism research: A review essay. Nonprofit and Voluntary Sector Quarterly, 41(2), 176–212.

77.

Yang

Y. C.

Boen

Harris

K. M.

(2015). Social relationships and hypertension in late life: Evidence from a nationally representative longitudinal study of older adults. Journal of Aging and Health, 27(3), 403–431.