This review summarizes the current knowledge about volunteering in later life and suggests 5 research questions at the forefront of knowledge development.
Rates of volunteering do not decline significantly until the middle of the 7th decade, and older volunteers commit more hours than younger volunteers. Older adults with more human and social capital tend to volunteer, and there is good evidence of a reciprocal relationship between volunteering and well-being. Program and policy developments in the field are outstripping production of knowledge to support evidence-based practices.
Research on the dynamics of volunteering over the life course as well as the patterns of activities that co-occur with volunteering is needed to guide program development. Research methods and findings from transdisciplinary work on the mechanisms through which psychosocial conditions affect health must be extended to the study of the effects of volunteering on older adults. Finally, we need to engage in more applied social science aimed at improving volunteer management, especially recruitment and retention of older volunteers.
VOLUNTEERING has historically been one of the few formal roles available to older adults after exiting the workforce as well as a clear-cut strategy to maintain the involvement promoted by activity theory. Indeed, volunteer activity fit well with the “busy ethic” that shaped modern retirement (Ekerdt, 1986). More recently, the discussion has changed: civic engagement is a topic of serious interest. The new-found interest was sparked by Putnam's (1995) analysis of the decline of civic participation over the last several decades, and his call for reversing these trends by restoring civic engagement. Although youth remain the primary target of civic engagement programming, there is increased attention to volunteering in later life. This interest stems from the compelling idea that volunteering not only strengthens civil society, it simultaneously improves the lives of older adults—perhaps more than younger people. Advocates point to the growing human capital of the aging population and call for expanding volunteer involvement—for the sake of communities as well as older adults themselves. Research on the topic has grown substantially along with this advocacy movement.
This review article focuses on formal volunteering. Although civic engagement is a wider concept that includes membership in voluntary associations and political participation, volunteering has dominated discussions of civic engagement and drawn the most attention from researchers. We define volunteering as an activity undertaken by an individual that is uncoerced, unpaid (or minimal compensation to offset costs), structured by an organization, and directed toward a community concern (Cnaan, Handy, & Wadsworth, 1996). This definition does not include helping others outside of the auspices of a formal organization (informal volunteering), caregiving, or making financial contributions. These are very important activities for society, and they have effects on the older adults who engage in them. However, they are distinct from formal volunteering and deserve more attention than possible in this review. Furthermore, international research on volunteering is growing, and how sociopolitical contexts affect formal volunteering is an important topic. This review focuses largely on volunteering in the United States.
Older adults engage in a wide range of formal volunteer activities that vary on several dimensions, including intensity and duration of commitment, type of organizational sponsor, type of service activity, the extent to which volunteers interact with staff and other volunteers, and organizational supports provided. There are episodic opportunities, such as serving a Thanksgiving meal at a homeless shelter. There are time-intensive but time-limited opportunities, like traveling to another community to build houses for Habitat for Humanity. There are the federal service programs, like Foster Grandparents or Senior Companions, which are highly structured and require a major commitment (volunteers serve for 15–40 hr/week and receive a small stipend of approximately $2.65/hr). This variety challenges the systematic development of knowledge about volunteering, and Musick and Wilson (2008) point out that the taxonomy of volunteering is in its infancy and no clear social science guidelines guide research.
Scholars have amassed a great deal of descriptive knowledge about volunteering. This article begins with a overview of what is known about rates of volunteering and associated factors. Then, five research questions at the forefront of knowledge development are reviewed.
BACKGROUND ON VOLUNTEERING AMONG OLDER ADULTS
Rates of volunteering calculated from self-reported surveys are inflated because individuals who agree to participate in surveys are more likely to be volunteers (Abraham, Helms, & Presser, 2009). Despite this limitation, the U.S. Bureau of Labor Statistics (2009) estimates from the Current Population Survey that 23.5% of adults aged 65 years and older volunteered in 2008 compared with the higher rates of 31.5% for 35- to 44-year-olds and 30.8% of 45- to 54-year-olds. The rate of volunteering among older adults has increased steadily for three decades, and older volunteers put in more time than younger volunteers (Foster-Bey, Grimm, & Dietz, 2007). The median time commitment for older adults is 90 hr/year compared with 52 hr/year for those 45–54 years and 60 hr for those 55–64 years. Forty-five percent of the older volunteers serve more than 100 hr/year, with 14% volunteering less than 15 hr and 21% between 15 and 49 hr annually (U.S. Bureau of Labor Statistics). The median time of 90 hr/year could be interpreted as close to 2 hr/week. However, there are many services that are not dispersed evenly across the year—like helping with income tax preparation or serving in a 9-month academic program. The large variation in volunteer hours and time frames in which the hours are completed challenge generalizations about the volunteer experience.
There are several explanations for the lower rates of volunteering among older cohorts. Younger adults take on volunteer roles related to their children's activities and their work roles. Older adults are more separated from educational and work institutions and are thus less likely to be presented with volunteer opportunities. Older adults are less likely to be asked to volunteer and being asked is a major route to volunteering (Independent Sector, 2000). Also, health issues are related to declining rates of volunteering for those over the age of 75 (AARP, 2003). However, it is important to note that educational levels of today’s older adults are lower, and some of the cohort differences in volunteer rates disappear when educational level is controlled (Musick & Wilson, 2008).
Older adults with more education, income, health, social integration, and religious involvement are more likely to volunteer (Tang, 2006; Wilson & Musick, 1997; Zedlewski & Schaner, 2006). There are differential rates of volunteering among ethnic groups, with older adults of color volunteering at lower rates than Whites. This may be related to historic segregation, disparities in economic and health resources, as well as structural barriers related to discrimination (McBride, 2007).
Current cohorts of older volunteers differ from younger volunteers. Older adults are more likely to volunteer for religious organizations and health and social/community service agencies and less likely to volunteer for educational, recreational, and environmental programs (Musick & Wilson, 2008; U.S. Bureau of Labor Statistics, 2009). Older adults are more likely to be involved in relational activities, like being a tutor, mentor, or friendly visitor (Morrow-Howell, 2007). Research largely suggests that older adults give different reasons than younger adults when asked why they volunteer (Omoto, Snyder, & Martino, 2000). In general, younger and middle-aged adults are more motivated by instrumental purposes (developing knowledge/skills and advancing careers), whereas older adults are more motivated by the desire to help others and to stay active (Okun & Schultz, 2003).
Several theories have been the mainstay of research on volunteering. Continuity and activity theories have been used to explain substantial rates of volunteering among older adults, with a life course perspective informing changes in volunteer involvement over time. Role theory has been used to explain participation, using concepts of role loss, role identity, and role continuity. A resource perspective has been used to explain who comes forward to volunteer, with higher levels human, social, and cultural capital leading to volunteer involvement. Motivations to volunteer in later life have been tied to developmental tasks of achieving generativity and leaving a legacy. More recently, age-related differences in volunteering have been explained with socioemotional selectivity theory. By and large, these theories have been used to explain empirical patterns, with less attention devoted to testing theory directly.
Within the growing literature on older volunteers, critical gerontologists have pointed out that formal volunteering has been dominated by individuals with more resources (Martinson & Minkler, 2006; Minkler & Holstein, 2008). With the spotlight on volunteering, older adults of color and those with fewer resources may be judged on their participation and formal volunteering may overshadow informal volunteering and caregiving, other productive activities critical to society. Also, there is concern that older adults are being called on as substitutes for necessary public support to nonprofit and public organizations. As we go forward on the topics reviewed subsequently as well as other topics related to civic engagement and the aging population, it will be useful to bear these critiques in mind.
LEADING-EDGE QUESTION 1: WHAT ARE THE “DYNAMICS” OF VOLUNTEERING AMONG OLDER ADULTS?
As seen in the above review, most descriptions of volunteering are static—who volunteers, for how much time, for what reasons, and with what organizations. Yet volunteering is a dynamic process. Participation starts and stops, waxes and wanes in response to changes in individuals’ lives as well as in response to the nature of the volunteer service. Like other areas of aging research, the dynamics may be different for those volunteers who grow old versus older adults who become first-time volunteers.
Rotolo (2000) studied transitions into and out of voluntary associations (although this work did not focus on older adults) and documented that volunteers leave one organization for another. This movement is not captured by measurement approaches, which record whether an individual volunteered during a certain time period. Movements might reflect individual preferences or constraints, they may be associated with structural characteristics of the volunteer work (an episodic service is completed), or they may relate to organizational arrangements that are mutable. Clearly, understanding these transitions is important to organizations seeking to recruit and retain volunteers.
Butrica, Johnson, and Zedlewski (2009) proposed that older individuals start or stop volunteering because of costs and benefits associated with these actions. They hypothesized that changes in health, work, marital status, and so forth, altered the costs and benefits of volunteering. They found that volunteering among older adults was mostly stable: Those who volunteered at baseline did so across the 8-year observation period, and those who were not volunteering at baseline did not start. They showed that the same variables affected both starting and stopping but that the effects were smaller in regard to starting. They concluded that it may be easier to influence volunteers staying (or not quitting) than it is to get them to start. Using data from 11 European countries, Hank and Erlinghagen (2010) came to a similar conclusion: The same factors have reverse effects on taking up and giving up volunteering. They also demonstrated that volunteer dynamics were influenced by varying sociopolitical contexts.
Li and Ferraro (2005, 2006) studied who selects into volunteering and the effects of volunteering on depression. They documented that older adults with higher depressive symptoms sought volunteer activity, perhaps as a way to improve mental health. Older adults with more physical health problems did not enter into volunteer roles. Continued engagement in volunteering was related to reduced depressive symptoms. Furthermore, their analyses revealed that nonrandom attrition of volunteers had serious effects on estimates between participation and depression. Volunteers and those with less depression were more likely to remain in the sample over time, and when accounting for this sample bias, the effects of volunteering were attenuated.
The wider availability of longitudinal data and statistical techniques to analyze dynamic processes will improve the study of volunteering. We are moving beyond understanding individual characteristics associated with volunteering to clarifying how different life experiences relate to volunteering, especially transitions in working and caregiving and changes in income, marital status, and health associated with later life. More nuanced understandings of what influences older adults to start, continue, or quit volunteering are important to organizations that depend on volunteers.
LEADING-EDGE QUESTION 2: WHAT IS THE RELATIONSHIP OF VOLUNTEERING TO OTHER SOCIAL ACTIVITIES?
In general, researchers focus on a single domain of activity, like volunteering, caregiving, or working, ignoring the reality that people engage in multiple activities. Patterns of activities are probably better predictors of outcomes than single activities, and the extent to which activities complement or compete with each other may be important in understanding volunteer involvement.
One approach to understanding concurrent involvement involves aggregating the number of roles undertaken by an individual. For examples, Adelmann (1994), Baker, Cahalin, Gerst, and Burr (2005), and Hinterlong, Morrow-Howell, and Rozario (2007) summed the number of roles assumed by older adults (including caregiver, employee, volunteer, etc.) and documented that more positive well-being was associated with more role involvements. Bukov, Maas, and Lampert (2002) suggested that social participation is cumulative, meaning that more resource-demanding forms of participation (political or civic engagement) are undertaken in addition to less demanding forms of engagement (social engagement in groups/collective group action).
Researchers have looked specifically at certain roles in combination to understand the extent to which they compete or complement each other. Most of this work has focused on the co-occurrence of paid work and caregiving (Moen, Robison, & Fields, 1994; Pavalko & Artis, 1997), but volunteering and caregiving as well as volunteering and work have been studied.
Cross-sectional descriptions suggest that employed adults volunteer more than unemployed, and the part-time workers have the highest volunteer rates (Choi, 2003). A 2007 study by the Corporation for National and Community Service (Foster-Bey et al., 2007) showed that the rate of volunteering among baby boomers declined among those who left the labor force or became empty nesters, suggesting the importance of institutional ties. Yet, the longitudinal picture of volunteering as it relates to paid employment is complicated, and Musick and Wilson (2008) suggested that it not retirement per se, but other factors associated with retirement that relate to drops in volunteering rates. Musick and Wilson's work showed that fully retired people put in more volunteer hours than those who work part-time in retirement as well as those who are not retired.
Volunteer behavior before retirement is a critical factor in the effects of retirement on volunteering (Chambre & Einolf, 2008). Mutchler, Burr, and Caro (2003) documented that changes in work status did not affect volunteer retention, but stopping work or stepping down to part-time was associated with starting volunteering. The interconnection of work and volunteering suggests the potential role of the business sector in promoting volunteering in the aging population. The role of the workplace was discussed at the 2005 White House Conference on Aging, in terms of provision of subsidies, tax credits, and other incentives to encourage the business community to expand volunteer opportunities for their retirees.
Burr, Choi, Mutchler, and Caro (2005) focused on caregiving and volunteering. They found that caregivers were more likely to be asked to volunteer than noncaregivers. This work suggested that caregivers may be motivated to help others through both informal and formal helping activities and that caregivers were more likely to be embedded in social networks that foster volunteering. In general, there is no clear evidence that caregiving stands in the way of volunteering (Musick &Wilson, 2008). Yet, Choi, Burr, Mutchler, and Caro (2007) focused on spousal caregivers and found that for women, but not men, caregivers were less likely to engage in formal or informal volunteering. The circumstances of the caregiving and the care recipient are likely important factors in the extent to which caregiving competes with volunteering.
Recent work has moved beyond studying the relationship of two activities to looking at combinations of multiple activities. Burr and colleagues studied co-occurring productive behaviors, including formal volunteer work, informal helping, unpaid domestic work, caregiving, and paid work, and identified four categories of older adults—helpers, home maintainers, worker/volunteers, and super helpers. They concluded that activities did not occur independently but were patterned and that productive activities were more likely to complement than compete with each other (Burr, Mutchler, & Caro, 2007). In another example of the simultaneous consideration of multiple activities, Sugihara, Sugisawa, Shibata, and Harada (2008) tested each of three activities (paid work, unpaid work at home, and volunteer work) in relation to depression. In a simultaneous regression model, they entered all possible combinations of these activities, including none. They found that involvement in one productive role may be more important than multiple roles and that volunteering was protective against loss of paid work.
Clearly, volunteering does not happen in isolation of other activities, and volunteering is only one of many important activities for society. Perhaps, the balance of activities matters most for the individual. For example, Hao (2008) found that, among adults aged 55–66 years, the pattern of working full time while volunteering part-time was the most protective of psychological well-being. To date, most studies of co-occurring activities have focused on productive activities, excluding leisure, religious, or social activities. Yet these activities are likely important in the balance that maximizes outcomes for the individual. The empirical issue of how to assess and analyze multiple activities and patterns remains a challenge. Methodological advancements are needed to address the important topic of how people balance work, volunteering, caregiving, leisure time, and so forth, and with what outcomes.
LEADING-EDGE QUESTION 3: UNDER WHAT CONDITIONS DOES VOLUNTEERING ENHANCE THE WELL-BEING OF OLDER VOLUNTEERS?
A long list of well-being outcomes have been associated with volunteering through longitudinal studies: reduced mortality (Musick, Herzog, & House, 1999), increased physical function (Lum & Lightfoot, 2005; Moen, Dempster-McClain, & Williams, 1992), increased levels of self-rated health (Luoh & Herzog, 2002; Morrow-Howell, Hinterlong, Rozario, & Tang, 2003), reduced depressive symptoms (Musick & Wilson, 2003), and increased life satisfaction (Van Willigen, 2000). Clearly, selection is an issue in this work, as older adults with higher levels of well-being volunteer. The work of Thoits and Hewitt (2001), Li and Ferraro (2005), and Hao (2008) documented that both forces operate—that is, older adults with higher levels of well-being volunteer and also experience positive effects because of volunteer participation.
In general, the evidence regarding the effects of volunteering on well-being outcomes derives from longitudinal analyses of secondary data sets, including the Health and Retirement Study, Americans’ Changing Lives, MidLife in the United States, and Longitudinal Study of Aging. These data sets have large representative samples and a wealth of standardized assessments of individual characteristics to use as both predictors and outcomes of volunteering. Analytic techniques take advantage of the multiple waves of data to study the effects of volunteering on subsequent outcomes. However, there are limitations to this approach: information about volunteering is very gross (whether or not the person volunteered in the observation period, for how much time, and for what types of organizations); a wide range of volunteer activities is included; and the nature of the work, the organizational supports, and many other relevant factors about the volunteer experience remain unspecified. Thus, it is not clear what constitutes the health-promoting intervention, and specifying causation remains difficult. In sum, the current literature presents a convincing argument that volunteering produces higher levels of well-being, but we are hampered in developing programs and policies without more specific information about what types of volunteer programs produce what kinds of effects—and under what circumstances.
There is some research moving in these directions. For example, outcome researchers have recently focused on Experience Corps (EC) as a model civic engagement program for older adults. EC is a national program where older adults are specifically recruited and trained to work with students and teachers to improve academic outcomes. Features of the program have been described elsewhere and in enough detail for specifying the intervention (Fried et al., 2004; Morrow-Howell, Jonson-Reid, McCrary, Lee, & Spitznagel, 2009). Furthermore, experimental and quasi-experimental designs are being implemented to document outcomes for the older volunteers. One-year outcomes showed that EC volunteers in Baltimore increased physical strength, increased in the number of people they could turn to for help, watched less TV, and showed less decline in walking speed, and there was a trend toward improved cognitive function compared with a waitlist comparison groups (Carlson et al., 2008; Fried et al., 2004). In a quasi-experimental design, Hong and Morrow-Howell (2009) documented that EC participants from 15 program sites across the country experienced improvements in depressive symptoms and functional limitations after 2 years of service compared with a matched-comparison group. Collectively, these studies are connecting outcomes more directly to intensity and duration of service, type of service, and program characteristics. This type of work may contribute to the description and dissemination of evidence-based volunteer programs.
Both the conditions of the older volunteers and the conditions of the volunteer experience need to be considered when documenting outcomes. The studies of volunteering and outcomes based on longitudinal analyses of secondary data sets have used interaction terms to test the moderating effects of individual characteristics on the relationship of volunteering and outcomes. There is some empirical support for the idea that subgroups of older adults differentially benefit from volunteering. Those with fewer personal and social resources may benefit the most (Morrow-Howell, Hong, & Tang, 2009; Musick et al., 1999; Piliavin & Siegl, 2007). Greenfield and Marks (2004) found that individuals with more role losses (those without social roles as spouses, employees, and parents) gained more from volunteering. Yet other studies documented that older adults who were married, employed, and had higher levels of religious and social involvement experienced more benefits (Oman, Thoresen, & McMahon, 1999; Van Willigen, 2000). A clearer understanding about differential effects of volunteering could guide program development around targeted recruitment to increase health outcomes as well as provide insights into the causal relationship between volunteering and health.
Just as individual volunteers vary a great deal in their characteristics, the volunteer experiences vary greatly as well. The nature of the work, level of participation, and organizational supports likely affect volunteer outcomes and perhaps more than characteristics of the volunteers themselves (Morrow-Howell, Hong, et al., 2009). There is evidence that a curvilinear relationship exists between level of participation and outcomes. Musick and colleagues (1999) found that volunteers serving 40 hr/year or less had better mortality outcomes than those who participated more, and Van Willigen (2000) used the same data set to document that the positive effects of volunteering tapered off after 100 hr/year. Windsor, Anstey, and Rodgers (2008) found that maximum psychological benefit was associated with volunteering between 100 and 800 hr/year. Luoh and Herzog (2002) reported a threshold effect (volunteering more than 100 hr annually positively affected mortality). What happens during the time that a person is volunteering is important as well—the amount of social interaction, the extent to which the work is meaningful, and the relationship with staff and other volunteers. As an example of the type of research that is needed, Van Willigen provided evidence that older adults benefit most from volunteer work with religious-based organizations.
In sum, there is a general understanding about positive effects of volunteering on older adults. We need more clarity about specific conditions of volunteering that produce these positive effects. At this point, it appears that less advantaged older adults, in terms of social, economic, and health resources, benefit more from volunteering, but we do not have enough evidence to advocate for targeting certain groups to maximize volunteer outcomes. There is little specific research on program features that maximize outcomes. Such research would clearly inform the development of interventions that are most beneficial to older adults. Ultimately, researchers need to vary conditions of the volunteer experience (the amount and type of work, etc.) but this experimental methodology will be very challenging.
LEADING-EDGE QUESTION 4: WHAT ARE THE MECHANISMS BY WHICH VOLUNTEERING PRODUCES WELL-BEING FOR OLDER ADULTS?
Gerontological scholars have long believed that activity is good for older adults. Yet, in promoting civic engagement, Freedman (2001) proposed that all activity is not created equal. Indeed, the attraction to volunteering is the “win-win:” the social value to society and the personal benefit to older adults. Volunteering may have a more positive effect on older adults than younger ones (Van Willigen, 2000). Furthermore, volunteering may be more health producing than other types of social participation (Piliavin & Siegl, 2007). The development of volunteer programs that maximize outcomes requires better understanding of the mechanisms through which this particular activity produces positive effects on older adults.
Certain characteristics of the volunteer role may increase its positive effects compared with other social roles. Volunteering is socially valued, publicly recognized, and more discretionary than working or caregiving. Role enhancement and role accumulation have been used to describe the increases in personal and social resources associated with volunteering. It has been theorized that the volunteer role is especially salient in later life as other social roles are generally being lost (Greenfield & Marks, 2004; Hao, 2008). Li (2007) found that widows who added a volunteer role after spousal loss were protected against depressive symptoms, and those who increased volunteer hours after widowhood experienced gains in self-efficacy.
Scholars have also proposed that the altruistic nature of volunteering contributes to its health-promoting qualities. For example, Brown, Brown, House, and Smith (2008) studied the effects of helping behavior on recovery from spousal loss and found that bereaved individuals who engaged in helping others experienced a more rapid decline in depression than those who did not. Greenfield (2009) assessed one aspect of altruism, the felt obligation to help others, and documented its protective role again psychological losses in the face of functional decline. The concept of “mattering” has been studied in regard to volunteering. Piliavin and Siegl (2007) documented that mattering (operationalized as the extent to which others are aware of and rely on the individual) mediated the effects of volunteering on psychological well-being. They also found that volunteering, but not participation in other social organizations, positively affected outcomes, and they concluded that the beneficial effect of volunteering is related to its altruistic nature. Scholars are documenting the positive health effects of having a strong sense of meaning in life (Krause, 2009), and the extent to which volunteering increases one’s sense of purpose warrants close examination.
Hendricks and Cutler (2004) tested socioemotional selectivity theory (Carstensen, 1992) as it related to volunteering. They studied rates of volunteering, number of hours, and number of organizations across age groups and observed that older adults were less likely to volunteer and volunteer for fewer organizations but that the number of hours and number of organizations remained quite stable for those who did volunteer. They also observed that a logistic model fit the volunteer data best, with involvement increasing with age until leveling off and maintaining in older age groups. Their findings support the idea that older adults drop peripheral roles but maintain involvement in roles that are emotionally meaningful.
Fried and colleagues (2004) developed the social model of health promotion, suggesting that volunteering produces positive outcomes through activities associated with the volunteer role—that is, through physical, cognitive, and social pathways. Tan and colleagues demonstrated that the high-commitment volunteers in the EC program became more physically active and sustained increased activity levels after 3 years (Tan, Xue, Li, Carlson, & Fried, 2006; Tan et al., 2009). In the same parent study, a pilot project matched eight EC volunteers to nine matched controls, and functional magnetic resonance imaging showed changes in brain activity; the neural gains of volunteers corresponded to improvements in executive function (Carlson et al., 2009). In terms of social pathways, participation in the EC program, as in most volunteer programs, involves social activity, including the development of relationships with agency staff and other volunteers, group meetings, and socialization events. A long history of research demonstrates the link between social integration and health, via altered health behaviors, modulation of neuroendocrine reactivity, and improved immune function (Marmot & Wilkinson, 2006).
A compelling aspect of Fried's conceptualization about the health-promoting aspects of volunteering is the idea that the nature of the volunteer work pulls older adults into the activity and keeps them there long enough to experience positive effects. That is, older adults are motivated to participate to be “generative and make valued contributions to society” (Carlson et al., 2008, p. 799), not to participate in a health promotion program per se. Thus, volunteering reduces problems of retention associated with traditional health promotion program, like exercising (Tan et al., 2009).
Researchers on other topics clearly related to volunteering, like social integration, altruism, purpose in life, and positive emotional states, are seeking to clarifying relationship of psychosocial conditions and health outcomes. Downstream factors or more proximal factors like health behavior (exercise, help-seeking behavior), psychological conditions (self-esteem, self-efficacy), and physiologic states (allostatic load, immune function) are being tested (Berkman, Glass, Brissette, & Seeman, 2000) as causal mechanisms. Cutting edge research on volunteering needs to embrace transdisciplinary work to identify these factors. A big challenge will be identifying which aspects of volunteering—the social integration, positive emotional states, altruism, and so forth—are the essential elements. Multiple elements of volunteer activity and multiple mechanisms are likely involved and may account for research findings that volunteering is more health producing than other social activities. Advancing understanding about causal mechanisms can not only guide program development but also contribute to more general knowledge about successful aging.
LEADING-EDGE QUESTION 5: WHAT STRATEGIES ARE MOST EFFECTIVE IN VOLUNTEER MANAGEMENT, ESPECIALLY RECRUITINGAND RETAINING VOLUNTEERS?
Nonprofit and public organizations that rely on volunteers need an adequate supply of volunteers who are effective in the performance of assigned roles. Hager and Brudney (2004) outlined nine recommended volunteer management practices: supervision and communication, liability coverage, screening and matching volunteers to jobs, regular collection of information on volunteer involvement, written policies and job descriptions, recognition activities, annual measurement of volunteer impact, training and professional development for volunteers, and training for paid staff in working with volunteers. Their survey of 3,000 charitable organizations found that, while the best practices are widely known by organizations, they are not widely adapted. Regular supervision and communication with volunteers was the most widely adapted practice. Larger charities and those in the health arena were more likely to use recommended practices. Despite the importance of understanding the implementation and effects of these recommended practices, applied research in this arena is underdeveloped.
Recruitment and retention of volunteers has received the most attention from researchers. The most effective recruitment method is the “personal ask.” Older adults are less likely to be asked to volunteer, but those who are asked to volunteer do so at rates five times higher than those who are not asked (Independent Sector, 2000). Furthermore, Blacks are less likely than Whites to be asked (Musick, Wilson & Bynum, 2000). Volunteers ask others to join them. Thus, there is a reinforcing pattern when older adults or people of color are underrepresented in an organization and cannot recruit others. Musick and Wilson (2008) argued that participation rates may actually reflect rates at which certain populations are asked to volunteer. Disentangling who is asked to volunteer from who volunteers has important program applications. The importance of matching motivations to volunteer with recruitment messages has been demonstrated (Clary et al., 1998). Musick and Wilson (2008) found that African Americans rated all motivations to volunteer higher than Whites, despite lower rates of participation, suggesting that structural barriers suppress formal volunteering among African Americans.
Turnover of volunteers is costly to organizations, and about 30% of older adults drop out of volunteering after 1 year of service (Foster-Bey et al., 2007). It appears that turnover is associated with the nature of the volunteer work. For example, volunteer tenure was the shortest for those providing general labor or supply transportation and longest for those providing professional or management services (Foster-Bey et al., 2007). Organizations that had trouble recruiting volunteers also had trouble retaining them, and retention rates were better when current volunteers brought in new volunteers (Hager & Brudney, 2004). Other research suggests that volunteer turnover can be reduced by ensuring that the volunteers gain a sense of accomplishment and are recognized for their contributions, by effectively monitoring and supervising activities and by providing various cash or in-kind compensation to meet expenses (Cnaan & Cascio, 1999; Finkelstein, Penner, & Brannick, 2005).
An institutional perspective has been used to study recruitment, retention, and outcomes achieved among older adults (Hong, Morrow-Howell, Tang, & Hinterlong, 2009). Institutional capacity describes organizational arrangements, including dissemination of information; accessibility of the role; incentives provided for engagement; and the degree to which participation is facilitated through training, supervision, flexibility, recognition, and accommodation. These institutional characteristics may explain more variance in outcomes than individual factors (Morrow-Howell, Hong, et al., 2009). Higher levels of facilitation, an index comprised measures of training, supervision, assistance, recognition, flexibility, and stipends, were associated with more positive outcomes in terms of meeting volunteer expectations, amount of service, and perceived benefits of participation (McBride, Greenfield, Morrow-Howell, McCrary, & Lee, 2009). Low-income and non-White volunteers perceived facilitation as more important than higher-income and White volunteers (Tang, Morrow-Howell, & Hong, 2008). A particularly interesting area of study regards stipends, which represent the boundary between volunteering and paid work. Some argue that stipends undermine the altruistic nature of the service, whereas others argue that they are necessary to offset costs of volunteering that prevent wider participation. Some evidence supports the use of stipends to leverage inclusion of diverse populations, increase retention, and maximize the experience of older volunteers (McBride, Gonzales, Morrow-Howell, & McCrary, 2009).
Much of the scholarship on strategies to attract and manage volunteers is descriptive, often with unrepresentative samples of volunteers and volunteer organizations. Experimental designs are rare, and age is not specifically considered. In sum, there is a lack of rigorous applied research to guide the development and implementation of effective strategies. Interdisciplinary work, including partners from nonprofit management, human resources, communications, education, and human development, will be needed.
CONCLUDING THOUGHTS ON MOVING FORWARD
Research on volunteering is a growing interest area for gerontologists. More and better social science on the topic has emerged in the last 10 years. The work is advancing methodologically to more closely mirror the realities of volunteering—it is a dynamic process, influenced by individual, cohort, and life course factors, and it occurs in the context of many other activities that matter to society and to individual well-being. The research questions are becoming more nuanced, and conditions that modify the outcomes of volunteering for the organization and the individual volunteer are being specified. Theory is being tested more directly in attempts to increase understanding about why volunteering is health producing for older adults and why it may be more important to older adults than younger adults.
The current interest in civic engagement is exemplified by new federal initiatives and increased funding for program development by private foundations. Even though the value of older adults’ volunteer activity has been estimated at $44.3 billion annually (Johnson & Schaner, 2005), financial investment in research on volunteering among older adults has been limited. Most research is conducted on on-going studies not specifically designed to study volunteering, and primary data collection efforts fully addressing important questions on the topic do not achieve large representative samples. Public and private investment in the specific study of volunteering in later life is needed. Action in the civic engagement field is outstripping the developing knowledge base, and applied knowledge about volunteering in later life to guide program and policy initiatives has never been more important.
The author would like to acknowledge the helpful suggestions of the editor, Ken Ferraro, and two anonymous reviewers.