Volunteers’ initial motivations in a community-based end-of-life care programme and the associations with their eudaimonic well-being

ABSTRACT

Volunteers are integral to the success of community-based end-of-life care (EoLC) programme. However, volunteers’ exposure to death may affect their well-being and retention. This study investigated volunteers’ motivation to join EoLC teams, and their eudaimonic well-being. The self-determination theory (SDT) framed the analysis. A self-administered survey recorded demographics, the 30-item Chinese Version of the Volunteer Functions Inventory (C-VFI) and the 8-item Flourishing Scale (FS). Repeated measures analysis, MANOVA and multivariate regression models tested associations between motivation and eudaimonic well-being. Two hundred and eighty cases were applied for analysis. Eudaimonic well-being was positively associated with autonomous motivation and negatively associated with controlled motivation. To protect volunteers’ health, and optimise retention and EoLC service quality, volunteers’ motivations should be matched with tailored training and supports.

KEYWORDS:

• Initial motivation
• eudaimonic well-being
• volunteer
• end-of-life care
• SDT

Introduction

Volunteers provide invaluable and unpaid contributions to many community health programmes around the world (Morris et al., 2013; Riffin et al., 2016). Volunteers’ motivation and experiences whilst volunteering can significantly influence their ongoing participation (E. L. Deci & Ryan, 2008; Kim & Kim, 2019; Stukas et al., 2016; Y. Wu & Li, 2019). Thus, to optimise retention of volunteers in community programmes, it is important to understand their motivation, and recognise risks to their physical and mental health as a result of volunteering. This will inform strategies to protect volunteers from harm, ensure their ongoing commitment and boost the viability of under-funded community programmes (Zhou & Kodama Muscente, 2022).

Functional motives theory

Motivation is defined by Baumeister (2016) as ‘wanting’, arising from internal processes that drive behaviour (Reeve, 2018). Within the context of volunteering, motivation refers to the reasons individuals choose to participate and how they rationalise their decisions (Hustinx et al., 2010). There are different perspectives from which assess motivation. One is the symbolic approach, which views motivations as expressions of an individual’s values and beliefs, influenced by a broader social context (Hustinx et al., 2010). Another is the rational-choice perspective, which contends that motivation arises from a cost-benefit analysis, with behaviour driven by selective material incentives (Olson, 1965).

The third is the functional motives theory, which conceptualises motivation as a process through which individuals perceive volunteerism in terms of fulfilling various psychological needs (Clary & Snyder, 1999), which are termed as ‘functional motives’. Motivation thereof could be understood as the purpose of behaviour to fulfil a specified need. In accordance with the functional motives theory, organisations should gather comprehensive information regarding potential volunteers’ motivational factors during recruitment. This approach allows for better alignment between volunteering tasks and the underlying motives of volunteers, thereby enhancing overall engagement and satisfaction. Several scales in terms of functional motives theory, have been developed to measure volunteer function, such as the Volunteer Functions Inventory (VFI) and the Inventory of Motivation for Hospice Palliative Care Volunteerism (IMHPCV). IMHPCV though special for hospice service, lacked robust validity across diverse cultural contexts, and had only been tested on active hospice volunteers but not those freshly admitted ones, we decided to perform VFI to measure volunteers’ expectations of EoLC services.

VFI has been substantiated in various settings and populations (Chacón et al., 2017), which offers a broad range of functions for the publics who have not yet received EoLC training. This scale may provide a comprehensive understanding of volunteers’ psychological needs, enabling non-governmental organisations (NGOs) to align volunteering tasks with their motives. Nevertheless, the functional motives theory regards motivation as an ‘expression of pre-existing needs and dispositions, thus preceding the action instead of being constructed through (inter)action’ (Hustinx et al., 2010, p. 16); thereof it potentially overlooks contextual factors influencing motivation, by assuming motivation to be constant and independent of context without indicating the volunteering stages, which refers to experience in volunteering (for instance naïve volunteers, volunteers with limited experience or experienced volunteers) (Garbay et al., 2015; Planalp & Trost, 2009). Consequently, this leads to an ambiguous volunteer motivational profile. The extant studies indicated that value motivation was the major facilitator for volunteering (Chacón et al., 2017) and EoLC volunteers are depicted as those age over 50 retired women with strong altruistic motivation to help others in the western context (Morris et al., 2013). Yeun (2020) observed a similar profile in Korea. Nonetheless, these studies do not provide explicit insights into the volunteer stages linked to this particular motivational profile, which we believe may be influenced by the varying contexts of volunteering.

Furthermore, the functional approach, while valuable in understanding specific volunteer motivations, does not provide a comprehensive understanding about the relationship between motivation and various volunteering outcomes, such as satisfaction, well-being, and commitment. In this article, to clarify this phenomenon, we situated the timepoint at before training to measure motivation, that is, initial motivation, for those fresh volunteers admitted and introduce a new theoretical framework to solve gaps aforementioned.

End-of-life care volunteer’s eudaimonic well-being

Volunteering has been linked to numerous positive well-being outcomes, such as reduced mortality risk, fewer physical functioning limitations, increased physical activity, and improved psychosocial outcomes (Kim & Kim, 2019; Russell et al., 2022). Nevertheless, in EoLC experiences, the proximity to death may evoke psychological responses such as anxiety, fear, and a sense of powerlessness, potentially impacting both the physical comfort of service providers and the quality of service through its associations with mental health (French et al., 2017; Thiemann et al., 2015). Thereof, volunteers’ psychological well-being is imperative for screening before training. Ryan and Deci proposed that psychological well-being mainly constituted by hedonic well-being and eudaimonic well-being (Ryan & Deci, 2001). Eudaimonic well-being is defined as the actualisation of human potential, including personal growth, relatedness and sense of meaning in life (Ruini et al., 2003; Ryan & Deci, 2001). It could be used to understand the flourishing of human development. Given the emphasis on interpersonal relationships and collective well-being in Chinese culture, eudaimonic well-being is a suitable measure of Chinese psychological well-being (Tang et al., 2016; Veenhoven, 1991).

There is a variety of instruments to measure eudaimonic well-being, although not all are suitable for Chinese contexts. The Flourishing Scale (FS), a validated eudaimonic-centred instrument rooted in humanistic psychology that measures social-psychological prosperity and emphasises universal human psychological needs, such as competence, relatedness, and self-acceptance (Diener et al., 2010), has been tested in multiple studies, including the Chinese, which has demonstrated good reliability, factor structure, and convergent validity (Tan et al., 2021; Tang et al., 2016). It is a brief scale for measurement with only eight items about achievement in domains of relationships, self-esteem, purpose and optimism, which comprehensively addresses emotions and overall psychological well-being, and makes it valuable for researchers and practitioners across disciplines, we then chose it for this study.

VFI, eudaimonic well-being and self-determination theory

Previous literature suggests that volunteer motivation significantly influences well-being and retention (E. L. Deci & Ryan, 2008; Kim & Kim, 2019; Stukas et al., 2016; Y. Wu & Li, 2019). For example, Stukas et al. (2016) used VFI and a seven-item subscale of the Australian Unity Well-being Index to measure associations between motivation and well-being, which could predict volunteers’ intention to stay. However, to our knowledge, even if phenomenon of burnout and compassion fatigue are mentioned in death work, for instance, hospice volunteers have been reported to experience symptoms akin to burnout, including emotional exhaustion, depersonalisation, and low personal accomplishment (Slocum-Gori et al., 2013), few concerned the associations between eudemonic well-being and motivations in end-of-life care volunteers. The mechanism between the two are unclear.

Furthermore, although previous studies have drawn attention to the associations between motivation and well-being (Stukas et al., 2016), the exploration of connections between the Volunteer Functions Inventory (VFI) and well-being has revealed limitations in elucidating the relationships between motives and health. This lack of clarity extends to understanding why certain motives exhibit positive correlations with well-being, while others demonstrate negative associations (Güntert et al., 2016). To address the limitations of functional motives theory in explaining the association between motivation and well-being, we applied the self-determination theory (SDT) to better understand the relationships between initial motivation and subsequent eudaimonic well-being in Hong Kong community-based EoLC volunteers.

Ryan and Deci developed SDT as a framework for investigating motivation and human development (R. M. Deci & Ryan, 2017). This theory incorporates the concept of eudaimonia, highlighting the importance of flourishing and vital functioning in human development. It proposed that social and cultural contexts will facilitate or undermine the motivations through the satisfactions of basic psychological needs (autonomy, relatedness, competence) that beneficial for well-being and self-regulations. SDT therefore emphasised ‘the contextual conditions of support conducive to thriving and what kinds of circumstances undermined it’ (R. M. Deci & Ryan, 2017, p. 81).

Autonomous and controlled motivations are key terms in SDT (R. M. Deci & Ryan, 2017). Autonomous motivation encompasses intrinsic motivation, integrated motivation, and identified motivation, while controlled motivation comprises introjected regulation and external regulation. Higher self-determination correlated with flourishing is linked to improved well-being, with autonomous motivations correlating positively with psychological health, and controlled motivations correlating negatively with it (Coleman & Walshe, 2021; De Clerck et al., 2022; E. L. Deci & Ryan, 2008; Y. Wu & Li, 2019). Therefore, the quality of motivation matters and through SDT, we could explain why some motivations leading to positive outcomes and otherwise.

The SDT alone cannot determine specific volunteering functions. To understand the effects of volunteering motivation on individuals’ well-being, and to help organisations to align volunteers’ expectations with their training and tasks, volunteer functions must be integrated with the SDT. Over the last decade, a number of studies have consistently shown the connection between SDT and VFI to provide a comprehensive understanding of volunteerism (Güntert et al., 2016; Ramos et al., 2016; K. Sheldon et al., 2016; K. M. Sheldon et al., 2022). Researchers have attempted to combine VFI with SDT’s autonomous-controlled continuum, finding VFI could be interpreted under the SDT framework. Studies have also shown that value and understanding motives are associated with self-determined motivation, while social, protective, enhancement, and career motives are linked to controlled motivation (Oostlander et al., 2013; Ramos et al., 2016).

K. Sheldon et al. (2016) compared SDT and VFI when attempting to understand the motivations of environmental activists. They showed that both theories contributed to explaining activists’ motivation, and by comparing measures from SDT and functional motives theory, they were able to demonstrate that volunteers with value-based motives (VFI) and autonomous motives (SDT) showed greater commitment and less burnout, than those with controlled and protective, social, career, and/or enhancement motives. Comparable results were observed in their study of food bank volunteers (K. M. Sheldon et al., 2022). By combining these theories, Güntert et al. (2016) showed that self-determined motivation (value and understanding) had positive effects on volunteering outcomes, while controlled motivation (enhancement, protective, career and social) had negative or non-significant relationships with the outcomes. Drawing from the existing literature on integration of VFI and SDT and considering the degree of self-determination on behaviour, we classify the six motivations from the Volunteer Functions Inventory (VFI) (value, understanding, enhancement, protection, career, and social) within the continuum of autonomous-controlled motivation. This classification allows us to more effectively examine the associations between motivation and volunteer outcomes, as illustrated in Figure 1.

Figure 1. The research model.

Grounded on the extant literature, this study thus tested:

Hypothesis 1

(H1): EoLC volunteers are those age over 50 retired women with a strong value motivation to help others.

Hypothesis 2

(H2): Autonomous motivations (value and understanding) are positively associated with flourishing.

Hypothesis 3

(H3): Controlled motivations (enhancement, protective, career and social) are negatively associated with flourishing.

Materials and methods

Study site

Hong Kong has the world’s longest average life expectancy, with projections that 35% of its population will be aged over 65 years by 2069 (C&SD, 2020). It has already been recognised that Hong Kong faces an ever-increasing demand for EoLC as its population ages. The Hong Kong Jockey Club End-of-Life Community Care (JCECC) project is an academic-community partnership that was initiated in 2016 to respond to wishes of terminally-ill people in Hong Kong to die within their community (Ho et al., 2013). It relies heavily on volunteers to provide EoLC under the direction of clinical staff.

Study design

Since 2016, the public aged at least 18 years are recruited annually, in collaboration with the JCECC team at the University of Hong Kong and social workers from four NGO partners. Applicants can apply online or in paper. They consented after reading an information sheet explaining volunteer tasks, capacity-building procedures, risks, benefits, and requirement to provide at least six months service in the EoLC programme. After registration, they complete a risk assessment and a follow-up face-to-face interview as part of the motivational screening process, conducted by experienced EoLC social workers, to identify people who may not be suitable to volunteer in an EoLC programme. Those people who passed the screening process were then enrolled into a volunteer training programme. They would receive a pretraining, posttraining, 6-month and 12-month follow-up questionnaire respectively. Variables such as EoLC competence, motivation, psychological well-being were assessed in the pretraining wave (T₀, 2019/2020/2021). Then, the posttraining wave (T₁, 2019/2020/2021) and 6-month follow-up wave (T₂, 2019/2020/2021) and 12-month follow-up wave (T₃, 2019/2020/2021) data were also collected. This study only used the pretraining wave data. In total, we had 453 volunteers enrolled for the training programme from 2019 to 2021, but only 280 with complete responses on demographic, motivation and well-being variables were included for analysis. A detailed description of screening and recruitment procedures is reported elsewhere (Wang et al., 2020).

Ethics

This study was approved by the Human Research Ethics Committee of the University of Hong Kong (EA1902032).

Study measures

A self-administered questionnaire collected information on volunteers’ demographic, motivation and eudaimonic well-being.

Demographic variables

Age, gender, marital status, educational level, employment status, and religious beliefs were recorded once, during the JCECC programme volunteer registration process. Multiple-choice questions assisted in efficient demographic data collection, ease of administration and analysis. We recoded demographic variables for regression analysis: gender (male = 0, female = 1), age (under 49 = 0, 50 or above 50 = 1), marital status (single = 0, in relationship = 1), educational level (under college level = 0, college or higher level = 1), employment status (unemployed = 0, employed = 1), religion (without religious belief = 0, with religious belief = 1).

Eudaimonic well-being

We employed the Flourishing Scale to measure volunteers’ eudaimonic well-being. The Chinese version is simple to administer and demonstrates good reliability, factor structure and convergent validity (Tan et al., 2021; Tang et al., 2016). The Mandarin version of the scale has been validated in Mainland Chinese contexts (Tang et al., 2016), while a Cantonese version has been translated and used in a Hong Kong study (Chan & Ma, 2015). The Cantonese version was used for this study (Cronbach’s α coefficients of 0.916). Scale scores range between 7 and 56, with higher scores indicating better personal eudaimonic well-being.

Motivation

We chose the Chinese Version of the Volunteer Functions Inventory (C-VFI). The VFI was originally developed by Clary et al. (1998) in the United States and later translated and validated in Hong Kong by J. Wu et al. (2009). The VFI assesses six motivations: value (altruistic and humanitarian concerns for others); understanding (the desire to learn or exercise knowledge and skills that may otherwise remain unused); enhancement (psychological development); career (the pursuit of obtaining career-related experience); social (the aim to strengthen or create social relationships and address concerns over social rewards and punishments); and protective (the reduction of negative feelings or personal problems) (Clary et al., 1998). Each dimension is represented by five items in the inventory (total 30 items), rated on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). This scale has been tested in a range of settings and populations (Chacón et al., 2017), and it has been widely employed to align volunteers’ motives with specific benefits gained from volunteering (Güntert et al., 2016). Wu reported that the reliability in the Chinese context of the six dimensions ranged from 0.70 to 0.91 (overall Cronbach’s α coefficient 0.954), indicating that the C-VFI was applicable to the Chinese population. This scale was appropriate for our study as it describes a broad range of motivations relevant to Chinese volunteers who have yet to receive EoLC training.

Analytical strategy

Data analysis was conducted using SPSS software version 28.0, with alpha of p < 0.05 (2-tailed). Participants’ health status was included in regression analysis. The multivariate equivalent of repeated measure analysis examined differences within types of motivations. Multivariate analysis tested ‘different levels of the repeated factor (in this case, motivation) as different dependent variables’ (Maurissen & Vidmar, 2017). MANOVA was employed to explore differences in demographic profiles related to volunteers’ initial motivation. Multivariate regression analysis measured associations between initial motivation and subsequent eudaimonic well-being.

Results

Demographic characteristics of participants: Approximately 50% volunteers were younger than 50 years of age (51.1%), single (54.6%), unemployed (59.3%) and possessed religious beliefs (62.1%). The majority was female (78.2%), and had attained at least a college education (72.9%).

Initial motivation and eudaimonic well-being of participants

Mean motivation scores showed that value motivation was ranked the highest, followed by understanding, enhancement, career, protective, and social motivations. The rank ordering of motivation in EoLC volunteers was similar to general volunteers (Chacón et al., 2017; Zhou & Kodama Muscente, 2022). However, there were significant differences between value and other motivations, except for understanding (Table 1). This suggests that EoLC volunteers are primarily driven by altruistic values and a desire to learn (Clary et al., 1998; Morris et al., 2013). Having higher scores in understanding motivation showed a higher demand on acquiring EoLC knowledge within the context of the ageing society (Vanderstichelen, 2022).

Table 1. The means of motivations at baseline.

Motivation	Mean	Std. Deviation	t	df	p
Value	5.358	1.039
Understanding	5.294	1.100
Enhancement	4.034	1.489
Career	3.621	1.562
Protective	3.378	1.410
Social	3.009	1.615
Value-Understanding	0.064	0.908	1.185	279	0.237
Value-Enhancement	1.324	1.382	16.024	279	<0.001
Value-Career	1.736	1.586	18.316	279	<0.001
Value-Protective	1.979	1.384	23.930	279	<0.001
Value-Social	2.349	1.642	23.935	279	<0.001

The average Flourishing Scale (FS) score was 46.918 (SD = 5.077). There was a negative association between gender and flourishing (Table 2), whilst value and understanding motivations were positively correlated with flourishing. There were no other strong correlations between demographic variables and estimates of flourishing.

Table 2. Correlation analysis among study variables.

Variables	1	2	3	4	5	6	7
1 Value
2 Understanding	0.641***
3 Enhancement	0.448***	0.664***
4 Career	0.309***	0.551***	0.730***
5 Protective	0.393***	0.504***	0.798***	0.720***
6 Social	0.295***	0.355***	0.598***	0.628***	0.751***
7 Flourishing	0.283***	0.209***	−0.003	0.038	0.062	0.003
8 Gender	−0.102	−0.031	−0.162**	−0.155**	−0.169**	−0.050	−0.135*
9 Age	−0.151*	−0.202***	−0.125*	−0.118*	−0.043	−0.344***	0.046
10 Education	0.166**	0.162**	−0.008	0.069	−0.09	0.116	−0.046
11 Employment	0.034	0.095	−0.027	0.006	−0.057	0.071	0.021
12 Marital_status	0.006	−0.066	−0.042	−0.036	−0.001	−0.211***	0.082
13 Religion	−0.046	−0.088	−0.125*	−0.168**	−0.029	−0.241***	0.044
14 Physical Health	0.103	0.133*	0.037	0.042	−0.109	0.099	0.209***

*p < 0.05. **p < 0.01. ***p < 0.001

Demographic profile of initial motivation

Table 3 reports the impact of demographic factors (gender, age, education, marital status, employment status, and religion) on volunteers’ initial motivation. Significant overall effects were observed for gender, age, education, marital status, and religion.

Table 3. Demographic information of participants and initial motivational profile (N = 280).

Variables	N	%	Value	Understanding	Enhancement	Career	Protective	Social
			M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Gender
Male	61	21.8	5.557(0.949)	5.357(1.199)	4.472(1.533)	3.771(1.533)	3.810(1.491)	3.525(1.877)
Female	219	78.2	5.302(1.058)	5.270(1.071)	3.912(1.457)	3.581(1.516)	3.256(1.366)	2.874(1.503)
			F (3.416, 949.766) = 3.108, p = 0.018, η2 = .011
Age
Under 49	143	51.1	5.511(0.881)	5.480(0.883)	4.132 (1.484)	4.048(1.558)	3.474(1.475)	3.018(1.596)
Above 50	137	48.9	5.199(1.164)	5.099(1.263)	3.933(1.493)	3.177(1.442)	3.279(1.336)	2.999(1.641)
			F (3.355, 932.674) = 7.220, p < .001, η^2 = .025
Education
Under college	76	27.1	5.076(1.291)	5.003(1.311)	3.866(1.503)	3.326(1.466)	3.397(1.419)	3.247(1.598)
Above college	204	72.9	5.463(0.910)	5.396(0.991)	4.097(1.483)	3.732(1.586)	3.370(1.410)	2.929(1.611)
			F (3.437, 955.506) = 5.896, p < .001, η^2 = .021
Employment status
Unemployed	166	59.3	5.329(1.091)	5.207(1.193)	4.027(1.514)	3.530(1.611)	3.410(1.420)	3.084(1.634)
Employed	114	40.7	5.400(0.962)	5.409(0.937)	4.046(1.458)	3.756(1.486)	3.329(1.400)	2.916(1.579)
			F (3.414, 949.135) = 2.073, p = 0.093, η2 = .007
Marital status
Single	153	54.6	5.363(1.041)	5.352(1.033)	4.084(1.457)	3.923(1.577)	3.430(1.405)	3.024(1.600)
In relationship	127	45.4	5.351(1.041)	5.214(1.172)	3.975(1.530)	3.260(1.471)	3.313(1.418)	3.006(1.631)
			F (3.384, 940.651) = 4.993, p = 0.001, η2 = .018
Religion
Without religion	107	38.2	5.419(0.901)	5.417(0.998)	4.351(1.341)	4.099(1.448)	3.602(1.338)	3.068(1.571)
With religion	173	61.8	5.320(1.117)	5.217(1.155)	3.838(1.545)	3.326(1.561)	3.240(1.439)	2.972(1.646)
			F (3.399, 944.883) = 5.631, p < .001, η2 = .020

N = 280. N = number of respondents; % = percentage.

Table 4 indicates gender differences in enhancement, protective, and social motivations (p < 0.01), with males scoring higher in these dimensions, which was consistent with other studies (Musick & Wilson, 2007; Zhou & Kodama Muscente, 2022). No significant gender differences were found in value, understanding, and career motivations. Regarding age effects (Table 4), post hoc tests identified age-related differences in value, understanding, and career motivations (differentiating between participants younger than 50 years, or 50 years or older), where younger participants scored significantly higher.

Table 4. Post hoc pairwise comparisons on motivations.

Initial Motivations	Δ M-F	Δ Ya-Oa	Δ Ac-Uc	Δ Ue-Em	Δ Si-Ir	Δ Wr-Re
Value	0.255	0.312*	0.386**	−0.071	0.012	0.098
Understanding	0.082	0.380*	0.399**	−0.212	0.145	0.199
Enhancement	0.560**	0.199	0.231	−0.019	0.109	0.513**
Career	0.191	0.871***	0.405	−0.224	.662***	0.773***
Protective	0.552**	0.195	0.025	0.077	0.12	.362*
Social	0.660**	0.018	0.328	0.186	0.004	0.097

M: male; F: female; Ya: Young age; Oa: Old age; Uc: Under college; Ac: Above college; Ue: Unemployed; Em: Employed; Si: Single; Ir: In relationship; Wr: Without religion; Re: With religion.

Δ_M-F: mean difference between male and female; Δ _Ya-Oa: mean difference between young age and old age; Δ _Ac-Uc: mean difference between those education level above college and those under college; Δ _Ue-Em: mean difference between those unemployed and those employed; Δ _Si-Ir: mean difference between those single and those are in relationship; Δ _Wr-Re: mean difference between those without religious belief and those with religious belief.

Participants with at least a college education scored significantly higher in value (p < 0.01) and understanding (p < 0.01) motivations. Singles had significantly higher career motivation scores than those in relationships (p < 0.001). Participants without religious beliefs scored higher in enhancement (p < 0.01), career (p < 0.001), and protective (p < 0.05) motivations than people with religious beliefs.

Table 5 reports the findings of multivariate regression modelling which examined the initial motivations and their explanatory power on flourishing.

Table 5. Multivariate regression model for flourishing.

Variables	Flourishing
Value	1.305(0.310)***
Understanding
Enhancement
Career
Protective	−0.592(0.241)*
Social
Gender	−1.462(0.742)*
Education	−1.681(0.721)*
Health	1.509(0.447)***
Intercept	36.160
Adjusted R^2	0.143

Regression coefficients are unstandardised, and robust standard errors reported in parentheses. (Stepwise) *p < .05. **p < .01. ***p < .001.

The model had significant predictive ability (F = 9.289, p < 0.001), and explained 14.3% of the total variance. The marginal tests on the regression coefficients indicated that higher autonomous motivation (value) was positively associated with flourishing (β = 1.305, p < 0.001), and control motivation (protective) was negatively associated with flourishing (β = −0.592, p < 0.05), partially supporting the hypothesises.

Discussion

To our knowledge, this study is the first of its kind to test an integrated model of motivation and eudaimonic well-being in volunteers in a Hong Kong community-based EoLC programme. We confirmed the validity of the application of the integration model to the Hong Kong community-based EoLC context and provide a more comprehensive understanding of volunteer motivation. Our framework addresses the limitations of the functional motivation approach, which often overlooks the change of contexts to impact motivational profile. The integration of VFI and SDT sheds light on the varying strengths and directions in the associations between motivation and well-being, addressing a critical gap in the existing literature. Our study also provided an initial motivational profile of volunteers. This study then raised several issues for our concern:

The initial motivational profile of younger volunteers

Results indicated that younger volunteers with higher educational levels scored significantly higher in value and understanding motivation compared with all other subjects. Women without religious beliefs tended to volunteer to enhance self-growth and deal with their negative feelings; younger single people without religion were driven to volunteer in pursuit of career goals. Contrary to H1, our study found significant differences in the age of volunteers involved in EoLC when compared to research conducted by Morris et al. (2013) in the Western context and study of Yeun in Korean setting (Yeun, 2020). Specifically, the mean age of the current sample was lower than the average age observed in these studies, which was typically older than 50 years. Our finding challenges the prevailing assumption that EoLC volunteering is primarily undertaken by older individuals, as previous studies have consistently demonstrated a lack of younger volunteers in this domain (Planalp & Trost, 2009; Stelzer et al., 2018). However, it is possible that this difference in age composition may have been influenced by a sample size bias, as the recruitment of volunteers occurred during the COVID-19 pandemic. This may have reduced the availability of elderly participants who were more vulnerable to the virus, and less likely to volunteer. Nonetheless, our observations of volunteering post-Covid-19 in Hong Kong appear to support the ongoing presence of this phenomenon.

Another reason is related to the measurement timepoint. Many studies on EoLC volunteers have not addressed the stages of volunteering (Garbay et al., 2015; Planalp & Trost, 2009), and did not indicate in which status volunteers situated. It is possible that the recruitment profile of volunteers may change as younger individuals with a stronger motivation of gaining knowledge, substantiated by findings from earlier studies (Nissim et al., 2016; Planalp & Trost, 2009) as well as ours, were more likely to register as EoLC volunteers at the outset. However, younger volunteers may be less likely to make a long-term commitment, as highlighted in research by Cnaan et al. (2022) and Vanderstichelen (2022). The increase in average age in other settings then may be attributed to the retention of older volunteers after training. Thus, clarification of volunteering timepoints is important, as sample demographics may change after training.

The association between initial motivation and eudaimonic well-being in EoLC

SDT differentiates between autonomous motivation and controlled motivation based on the degree of self-determination exhibited in behaviour. Several studies have established positive associations between self-determination and well-being (Coleman & Walshe, 2021; De Clerck et al., 2022), and there are similar findings regarding the integration of SDT and VFI (Ramos et al., 2016; K. Sheldon et al., 2016). However, this integrated model has not yet been tested in the context of EoLC. Previous research has demonstrated that volunteers perceive EoLC service as meaningful and valuable (Morris et al., 2013), indicating that their involvement may not be solely driven by external norms or obligations. The higher autonomous motivation (value and understanding) observed in this study supports this notion.

Based on the theoretical assumptions of SDT, a positive main effect of autonomous motivation (value) on the Flourishing Scale was tested, revealing that autonomous motivation facilitates eudaimonic well-being. Furthermore, the results indicate that protective motivation, as a type of controlled motivation, is also significant in volunteers’ eudaimonic well-being. Volunteers with medium to high levels of controlled orientation may perceive substantial pressure and exhibit lower levels of flourishing, which partially confirms our second and third hypothesis. Still, in situations where autonomous motivation is strong while controlled motivation coexists in a weaker form, the impact of controlled motivation on psychological health requires further evaluation.

Implications

This integrated approach has its practical implications. It not only assists in matching volunteers with their tasks but also provides organisations with insight into the volunteers’ psychological conditions. By understanding the mechanism connecting motivation and well-being, organisations can potentially create opportunities for the transformation of volunteering motivation, thereby enhancing both volunteer satisfaction and organisational outcomes. For instance, if a volunteer enters the programme with strong protective motivation, that’s participating the training or service to avoid her/his own negative feelings, the agency can be alerted to help the volunteer releasing negative emotions and cultivating a connected context for personal growth. Therefore, while VFI can guide organisations on matching volunteers to suitable roles/activities that fulfil their volunteering expectations, integrating VFI with SDT enables a more person-centred approach to the volunteering process and contexts. This benefits not just patients but also the well-being of service providers.

Furthermore, about the implications from the change of volunteer motivational profile, our research concurs with the contemporary phenomenon of new volunteering in EoLC, suggesting that younger individuals are driven by values and understanding (autonomous motivation) while also experiencing considerable external pressures (controlled motivation). This insight informed the modification of our volunteer training approach.

Although initial interest in registering as an EoLC volunteer was high, few participants remained to engage in service provision following training, for instance, one of our partnered-NGOs reported that only around 4 out of 25 admitted volunteers stayed after training. This revealed that the acquisition of EoLC knowledge constituted a significant motivating factor for participants. In response to this challenge, we endeavoured to differentiate between those primarily seeking knowledge and those genuinely interested in providing service. Consequently, we developed an online self-learning platform accessible to all individuals interested in EoLC knowledge. For those committed to service provision, we required them to secure an affiliation with an NGO to gain opportunities for service provision. Subsequently, these individuals were eligible to receive advanced, trainer-led instruction to further enhance their EoLC skills. Whether this new strategy will affect the attrition rate deserves future observation.

This research design has limitations. Firstly, this study represents a preliminary attempt to integrate VFI and SDT in the context of EoLC services, which classify VFI motivations into autonomous and controlled categories. However, our results indicate that only value and protective motives had significant effects on the model, while the remaining four types of motivations were not found to have significant effects. These findings highlight the need for future research to further explore the contributions of the other four types of motivations to our understanding of the associations between motivations and EoLC volunteers’ well-being. A larger sample would be required to do this. Secondly, the measurement of eudaimonic well-being was limited in this study. In future research, additional indicators of EoLC volunteers’ well-being could be included in the analysis to provide a more comprehensive understanding of the relationship between motivations and well-being in the context of EoLC services. Additionally, comparative studies involving non-EoLC volunteers or volunteers in other healthcare settings could provide insights into the unique challenges and experiences faced by EoLC volunteers.

Acknowledgments

We thank our colleagues from the Jockey Club End-of-Life Community Care project team who provided insight and expertise that greatly assisted this study, and our partners at the Hong Kong Society for Rehabilitation, Hope of Heaven Christian Services, S.K.H. Holy Carpenter Community Centre, and St. James’ Settlement for their assistant with recruitment and training volunteers and providing valuable materials for data analysis of this study, together with those volunteers who participated in this project.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by The Hong Kong Jockey Club Charities Trust.

Notes on contributors

Zhuyun Lin

Dr. Zhuyun Lin is a post-doctoral fellow in the Faculty of Social Science at The University of Hong Kong. Her research interests lie in ordinary experiments, patients’ moral agency, caregiver’s motivation and mental health.

Vivian Weiqun Lou

Professor Vivian Weiqun Lou is the Director of Sau Po Centre on Ageing, and Professor at Department of Social Work & Social Administration at The University of Hong Kong. Her research interests focus on family gerontology, in particular family caregiving for dementia, stroke, end-of-life older adults, and social adaptation and mental health of Chinese older adults and family caregivers. She has committed to cross-sector collaboration with community partners, social enterprises, and the market. Prof. Lou has been appointed as a member of Statistics Advisory Board, Community Investment and Inclusion Fund, Senior Police Call Central Advisory Board, and Elderly Academy Development Foundation of the HKSAR Government. She is a fellow of the Gerontological Society of America, and a Council member of Hong Kong Association of Gerontology and Consortium of Institutes of Family Limited (CIFA). Professor Lou is recognized as Healthy Ageing 50 leaders transforming the world to be a better place in which to grow older in 2022 and Woman Change Maker in STEM in 2023. Recently, Professor Lou was recognized as 2023 Standford’s List of World’s Top 2% Scientists.