Sage Journals: Discover world-class research

Abstract

Episodic volunteers (EVs) are vital for non-profit organization activities. However, theory-based research on episodic volunteering is scant and the determinants of episodic volunteering are not well understood. This study integrates the volunteer process model and three-stage model of volunteers’ duration of service to explore determinants of EV retention. A cross-sectional survey of 340 EVs assessed volunteering antecedents, experiences, and retention. Social/enjoyment (β = .17) and benefit (β = −.15) motives, social norm (β = .20), and satisfaction (β = .56) predicted Novice EV (first experience) retention; satisfaction (β = .47) and commitment (β = .38) predicted Transition EV (2-4 years intermittently) retention; and supporting the organization financially (β = .31), social norm (β = .18), satisfaction (β = .41), and commitment (β = .19) predicted Sustained EV (5-6 years consecutively) retention. Integrated theoretical approaches appear efficacious for understanding EV retention. An Episodic Volunteer Engagement and Retention model is proposed for further testing in prospective work.

Keywords

episodic volunteering volunteer retention motives commitment

Volunteers are critical to the activities of non-profit organizations (NPOs). Of the 5.2 million Australians who volunteered in 2006, 84% contributed 623 million hr to the Australian non-profit sector with a wage equivalent value of Aus$15 billion (Productivity Commission, 2010). However, NPOs experience continued pressure to recruit new and retain existing volunteers in a changing landscape that includes declining median hours volunteered per person and an increasing demand for short-term, flexible, or “episodic” volunteering opportunities (Macduff, 2005). The net outcome of this change is reduced volunteer availability, high rates of turnover and increased costs for NPOs, many of whom do not have established programs to support episodic volunteers (EVs; McCurley & Lynch, 2005). Parallel to this, EVs are critical when large numbers of volunteers are needed over a short-time period (Macduff, 2004) such as during crises (Cnaan & Handy, 2005), to provide services to patients or specific community groups (Hustinx, Haski-Leventhal, & Handy, 2008), or for community events (Handy, Brodeur, & Cnaan, 2006). Hence, understanding and responding to the recruitment and retention of EVs is a crucial issue for NPOs and the volunteering sector.

Problematically, although episodic volunteering is a critical and growing phenomenon, empirical investigations about EVs are scant (Handy et al., 2006; Hustinx, 2005; Hyde, Dunn, Scuffham, & Chambers, 2014; Wilson, 2012). Moreover, research to date has focused predominantly on motives and/or satisfaction (Beder & Fast, 2008; Handy et al., 2006; Hustinx et al., 2008; Smith et al., 2010) rather than EV retention (Bryen & Madden, 2006; Cnaan & Handy, 2005). In addition, this research has seldom used theoretically driven approaches. One approach drawn from the traditional volunteering literature that may be of relevance is the Volunteer Process Model (Omoto & Snyder, 2002; Snyder & Omoto, 2008). This conceptual framework describes the key features of the volunteering process and structures these within three linked stages: antecedents, experiences, and consequences (Omoto & Snyder, 2002). These stages are proposed to span multiple levels encompassing individual, interpersonal, organizational, and societal systems. At the individual level, volunteer’s psychological processes (e.g., decision making) and behaviors are considered. Antecedent variables at this level include motives (Omoto & Snyder, 1995; Penner & Finkelstein, 1998), other’s expectations or social norm for volunteering (Finkelstein, Penner, & Brannick, 2005; Hyde & Knowles, 2013), and connection to or concern about the community (Omoto & Snyder, 2002). Experience variables incorporate satisfaction with volunteering (Omoto & Snyder, 1995; Penner & Finkelstein, 1998) and commitment to the organization volunteered for (Penner, 2002), and consequences variables include volunteer retention (Snyder & Omoto, 2008).

The Volunteer Process Model delineates particular features of volunteering that may be more relevant to a specific system level (e.g., motives at the individual level). However, one limitation is that it does not identify within a level when specific antecedent and experience variables may be most critical for traditional volunteer retention. For example, this model does not specify whether a commitment to the organization or its values exerts the strongest influence on volunteer retention after 6 months, 1 year, or 5 years. In their study of EVs at summer festivals, Handy et al. (2006) found significant differences in motives for volunteering between EV categories (based on frequency of participation), with “habitual” EVs and “long-term committed volunteers” being significantly more motivated to volunteer for altruistic (other-oriented) reasons compared with “genuine” EVs. Thus, it appears critical for models to give guidance regarding when in the episodic volunteering life cycle specific antecedent and experience variables may have the greatest impact on EV retention.

As a complement to the Volunteer Process Model, the Three-Stage Model of Volunteers’ Duration of Service (Chacón, Vecina, & Dávila, 2007) stipulates temporal estimates for when specific antecedents and experiences may be most influential on retention, and proposes intention to continue volunteering at each time-point as the main link between these variables and volunteering behavior. Specifically, motivation and satisfaction are proposed to predict intention to continue volunteering and subsequent volunteering behavior in the short term (≤12 months; Jiménez, Fuertes, & Abad, 2010; Vecina, Chacón, Sueiro, & Barrón, 2012). In the long term (>1 year) organizational commitment, rather than satisfaction, is proposed as the key predictor of volunteering intentions and behavior (Chacón et al., 2007; Vecina et al., 2012). One limitation, however, is that this model does not take into account the potential effects of volunteer setting or role on retention, factors which may vary widely for both traditional and episodic volunteering. To address the knowledge gap about EV retention, it is essential to understand not only the key antecedents, experiences, and consequences that drive continued episodic volunteering but also whether there are differences in these variables and EV retention based on role, setting, or phase in the episodic volunteering life cycle.

The Current Study

Accordingly, the present study aimed to explore the determinants of retention of EVs in a specific role and non-profit setting at 3 time-points in their experience: Novice EVs (volunteered for the first time), Transition EVs (volunteered sporadically for 2-4 years), and Sustained EVs (volunteered 5 or 6 years consecutively). In doing so, we adopt a novel, integrated theoretical approach to identify the key antecedents and experiences that may affect EV retention. These include motives for episodic volunteering, psychological sense of community (sense of connection the EV feels to the community in which they live), social norm (perceived support or pressure from others for episodic volunteering), satisfaction with episodic volunteering, and organizational commitment (loyalty to the organization, willingness to exert effort for the organization, and acceptance of the organization’s values). In accord with the Three-Stage Model of Volunteer’s Duration of Service, intention to continue volunteering was considered as a proxy for EV retention (Chacón et al., 2007), and these terms are used interchangeably in the present study. Within this exploratory study, we expected that motives, social norm, and satisfaction would predict intention to continue volunteering in the novice episodic volunteering phase whereas psychological sense of community and commitment would predict intention to continue volunteering in the transition and sustained episodic volunteering phases.

Method

Setting

In the current study, cancer control was chosen as one relatively homogeneous setting in which NPOs rely on the efforts of EVs via community events such as Relay For Life (RFL) to advance their mission. RFL, a global movement founded by the American Cancer Society, originated in the United States in 1985 with events now held in more than 600 communities across 20 countries, including Australia. The event honors people who have experienced cancer and their carers, and is the largest fundraising event for cancer internationally. In Australia, RFL events take place in every state and territory and raise more than Aus$24 million each year to fund cancer research, prevention, and support (Cancer Council, 2014). Teams of volunteers, including the team leader (i.e., Team Captain), fundraise prior to the event. This article reports data from volunteers who were Team Captains for RFL events in one Australian state in 2013. RFL events in this state were held over two seasons from April to June and July to November 2013 and attracted 24,196 episodic volunteers who formed 2,232 teams.

Participants and Procedure

Volunteers were eligible to participate in this cross-sectional study if they had registered their details on the event registration database and indicated they were willing to be contacted to receive event-related information. Initially, 1,546 volunteers were identified and contacted in November 2013 via email invitation (with accompanying information sheet and survey URL) to complete an online survey created using SurveyMonkey. A follow-up email invitation was sent 2 weeks after initial contact (December, 2013). Given that many events occur in regional areas where Internet access may not be optimal, a second follow-up was conducted 1 month later (January, 2014) to facilitate recruitment and comprised a mail-out of a hard copy invitation letter, information sheet, and survey. The information for 40 volunteers was outdated or incomplete. Of the remaining 1,506 volunteers contacted, 99 completed the electronic survey and 257 completed the mail survey (N = 356; 23.6% response rate). Ethical approval for this study was obtained from the Griffith University Human Research Ethics Committee.

Measures

Pilot survey

Prior to the main study, 6 staff involved in events and 20 event volunteers who participated in 2012 were invited to complete and comment on the survey and measures. Of these, four staff and three volunteers responded and confirmed completion time, suggested modifications to some items in the motives scale (e.g., to improve clarity) and face validity of the survey.

Demographics

Volunteers self-reported their gender, age in years, country of birth, marital status, education, employment, income, home ownership, length of residence in their community, and prior experience with cancer.

Duration of volunteering

Volunteers self-reported the specific years they were involved with the event in the period 2008 to 2013. Based on this information, three phases of episodic volunteering were formed: novice, transition, and sustained. These phases correspond broadly to Handy et al.’s (2006) proposed categories of genuine EVs, habitual EVs, and long-term committed volunteers, respectively. Given high rates of turnover associated with episodic volunteering (McCurley & Lynch, 2005), the first volunteer experience is a critical juncture at which to consider the determinants of intended future volunteering (i.e., retention) and thus formed the first episodic volunteering phase (novice). In the RFL context, the first experience is equivalent to 1 year duration of service. Following this first experience, EVs may return on an irregular basis (e.g., every few years) and in this transition phase begin to gain experience, and develop deeper connections to other EVs, the non-profit organization, or the cause itself. Therefore, a 2- to 4-year time frame comprised the second phase (Transition) in which the determinants of EV retention were examined. Finally, a core group of EVs sustain their volunteering and return to participate on a regular basis over an extended time. Thus, a 5- to 6-year period in which EVs participated consecutively formed the third phase (Sustained) in the current study.

Motives for volunteering

The scale developed by Won, Park, and Turner (2010) to examine motives for participating in events was adapted to form a 28-item scale for the current study and context (with 2 items removed from the original 30-item scale based on responses from the pilot study indicating the meaning of these items was unclear). Won et al. (2010) reported that this scale comprised six factors: Philanthropy (12 items), Social/Entertainment (5 items), External/Benefits (5 items), Family Needs (2 items), Sports (2 items), and Group Collaboration (2 items; Cronbach’s α ranged from .78 to .92). In the current study, all items were preceded by the statement “I volunteer for RFL because . . .” and responses to each item were rated on a 5-point Likert-type scale from 1 (strongly disagree) to 5 (strongly agree).

Psychological sense of community

A four-item scale adapted from Costa, Chalip, Green, and Simes (2006) and Nasar and Julian (1995) measured volunteer’s sense of connection to the community in which they reside. Good internal reliability has been reported for the larger 11-item psychological sense of community scale (α = .87; Nasar & Julian, 1995); however, in the current study the 4-item scale did not demonstrate strong internal consistency (α = .56). Removal of items improved reliability only marginally; therefore, the 4-item scale was retained. Example items were as follows: “I am quite similar to most people in my community” and “If there was a serious problem in the community, the people here could get together to solve it,” scored 1 (strongly disagree) to 5 (strongly agree).

Social norm

Volunteer’s perceptions of others’ expectations regarding their role as an event volunteer was measured with an adapted version of Callero’s (1985) Others’ Expectations eight-item scale (including three reverse-scored items; α = .85). In the current study, responses to each item were rated on a 5-point Likert-type scale, scored 1 (strongly disagree) to 5 (strongly agree). The scale was reliable with a Cronbach’s α of .84. Example items include the following: “It is important to my friends and relatives that I continue volunteering for RFL”; “No one would be really surprised if I just stopped volunteering for RFL.”

Satisfaction

Satisfaction with volunteering for the event was measured with the six-item satisfaction and personal fulfillment scale (Clary et al., 1998; α = .85) adapted to suit the context of this study. In the current study, items were scored on 5-point Likert-type scales from 1 (strongly disagree) to 5 (strongly agree), with excellent internal consistency of α = .92. Example items include the following: “My experience volunteering for RFL was worthwhile” and “I accomplished some ‘good’ through volunteering for RFL.”

Organizational commitment

Commitment to the state-based host organization was measured with the nine-item version of the Organizational Commitment Questionnaire (Mowday, Steers, & Porter, 1979). This scale comprised the positively worded items from the larger 15-item Organizational Commitment Questionnaire, and has a reported internal consistency of α = .95. In the current study, items were scored on 5-point Likert-type scales from 1 (strongly disagree) to 5 (strongly agree), and the wording “this organization” from the original scale was replaced with the name of the actual organization hosting the events. Example items included the following: “I talk up this organization as a great organization to volunteer for”; “I really care about the fate of this organization”; and “I am willing to put in a great deal of effort beyond that normally expected to help this organization be successful.” This scale demonstrated excellent reliability (α = .93).

Volunteer retention

An intention to continue volunteering scale developed by Garner and Garner (2011) was used in this study to measure volunteer retention. The scale was reported as having high internal consistency in prior work (α = .90) and consists of eight items that were modified to suit the context of the current study, and scored on 5-point Likert-type scales from 1 (strongly disagree) to 5 (strongly agree). Example items include the following: “I plan to volunteer for RFL in the future”; “I am more motivated to volunteer because of recent volunteer experience at RFL”; and “I will recommend that others volunteer for RFL.” The scale had excellent reliability in the current study with a Cronbach’s α of .94.

Data Analysis

Descriptive statistics (means, standard deviations, frequencies, percentages) were calculated to describe sample demographics and episodic volunteering phase (based on duration of service). Given that several scale items were modified to improve clarity and administered in a different geographic context, exploratory factor analysis using principal components analysis (PCA) extraction with varimax rotation was conducted to identify underlying dimensions of the motives scale. Multiple regression analysis on volunteer retention for each episodic volunteering phase was conducted with motives, psychological sense of community, social norm, satisfaction, and commitment as predictors.

Results

Descriptive Analysis of the Sample

Demographics and survey response type

Of the 356 EVs completing the survey, 16 respondents were removed due to substantial missing data giving a total sample of 340 EVs for analysis. Eighty-three (24.4%) completed the electronic survey and 257 (75.6%) completed the mail survey. Participants responding to the survey electronically or by mail did not significantly differ on the basis of their demographic characteristics, and therefore, the samples were combined for analyses. Volunteers ranged in age from 13 to 77 years with a mean age of 49.93 years (SD = 13.72). Most were female (88.8%), married (69.4%), born in Australia (84.1%), educated at high school level or above (97.6%), employed full-time (51.8%), earning less than Aus$80,000 (72.1%), paying a mortgage (49.7%), and residents in their community for more than 1 year (94.1%). Most of the sample knew someone who had been diagnosed with cancer (97.9%), and of these 80.6% knew someone diagnosed in the last 2 years.

Demographics and episodic volunteering phases

Eighty volunteers (23.5%) reported they had volunteered for the event for the first time and were classified as Novice EVs, 157 (46.2%) were Transition EVs having volunteered for the event 2 to 4 times, and 103 (30.3%) had volunteered 5 or 6 years consecutively for the event and were classified as Sustained EVs. Table 1 presents demographic characteristics for these volunteers overall and based on episodic volunteering phase. Potential differences in sample characteristics were explored. A one-way between-groups ANOVA confirmed that the average age between episodic volunteering phases was significantly different, F(2, 333) = 34.52, p < .001. Post hoc comparisons using Tamhane’s T2 (for unequal sample sizes) showed EVs in the sustained phase (M = 51.32, SD = 11.83) were significantly older than EVs in the novice (M = 37.20, SD = 13.09) and transition (M = 40.32, SD = 12.68) phases. The difference in average age between Novice and Transition EVs, however, was not significant. A greater proportion of EVs in the sustained phase owned their own home/were paying a mortgage compared with EVs in the transition and novice phases, χ² = 7.39, df = 2, p = .025. Compared with Novice EVs, EVs in the sustained phase reported more personal connections as having been diagnosed with cancer such as parent(s) (χ² = 11.50, df = 2, p = .003), a partner (χ² = 7.58, df = 2, p = .023), or sibling(s) (χ² = 8.19, df = 2, p = .017). Sustained EVs also were more likely than EVs in the transition phase to report a friend being diagnosed with cancer, χ² = 8.71, df = 2, p = .013. Novice EVs more often reported having one or more grandparent diagnosed with cancer than EVs in the sustained phase, χ² = 7.85, df = 2, p = .020. No other differences in sample characteristics based on episodic volunteering phase were significant.

Table 1.

Demographic Characteristics of Episodic Volunteers.

Characteristic	All EV N = 340	Novice EV n = 80	Transition EV n = 157	Sustained EV n = 103
Age in years
M (SD)	42.93 (13.72)	37.20 (13.09)	40.32 (12.68)	51.32 (11.83)
Gender (%)
Female	88.8	90.0	90.4	85.4
Male	10.3	8.8	8.3	14.6
Missing	0.9	1.3	1.3	—
Marital status (%)
Single	20.3	30.0	19.1	14.6
Married/de facto	69.4	63.8	70.7	71.8
Divorced/separated	7.6	5.0	7.6	9.7
Widowed	2.1	1.3	1.3	3.9
Missing	0.6	—	1.3	—
Country of birth (%)
Australia/New Zealand	85.9	91.2	86.6	80.5
Asia/Asia-Pacific	0.9	1.3	—	2.0
United Kingdom (England, Ireland, Scotland, Wales)	8.8	6.3	7.6	12.6
Europe	0.9	—	1.3	1.0
United States/Canada	0.9	—	1.9	—
Africa	2.0	1.2	1.3	3.9
Missing	0.6	—	1.3	—
Education (%)
University/college	39.4	46.3	40.1	33.0
Technical/trade	27.9	23.8	28.7	30.1
High school	30.3	28.8	28.0	35.0
Primary school	1.8	1.3	1.9	1.9
Missing	0.6	—	1.3	—
Employment (%)
Full-time	51.8	52.5	53.5	48.5
Part-time/casual	23.2	21.2	21.1	28.2
Full-time home/carer duties	5.9	5.0	7.0	4.9
Retired	5.6	2.5	4.5	9.7
Student	5.6	11.2	5.7	1.0
Other	7.0	6.3	7.7	6.8
Missing	0.9	1.3	0.6	1.0
Income in Aus$ (%)
<20,000	14.4	16.2	14.6	12.6
20,001-40,000	17.0	12.5	15.9	22.3
41,000-60,000	23.2	20.0	26.1	21.4
61,000-80,000	17.4	18.7	21.0	10.7
81,000-100,000	10.3	10.0	8.3	13.6
≥101,000	6.8	7.6	6.3	6.7
Unwilling to answer	9.7	12.5	7.0	11.7
Missing	1.2	2.5	0.6	1.0
Home ownership (%)
Renting	22.6	25.0	27.4	13.6
Owns home	20.9	13.7	16.6	33.0
Paying a mortgage	49.7	48.7	51.6	47.6
Unwilling to answer	5.9	11.3	3.8	4.9
Missing	0.9	1.3	0.6	1.0
Length of residence in community (%)
<1 year	5.0	5.0	7.0	1.9
≥1 year	94.1	93.8	92.4	97.1
Missing	0.9	1.3	0.6	1.0
Know someone diagnosed with cancer (%)
Yes	97.9	96.3	98.1	99.0
No	1.2	2.5	1.3	—
Missing	0.9	1.2	0.6	1.0
Person diagnosed (%)
Self	18.5	18.8	14.0	25.2
One (or both) parents	41.5	25.0	45.9	47.6
Partner	9.4	3.8	8.3	15.5
One (or more) children	4.4	3.8	3.8	5.8
Sibling (or more than one)	16.2	8.8	14.6	24.3
Grandparent	30.3	37.5	33.1	20.4
Other close family members	49.7	40.0	50.3	56.3
A friend	62.9	62.5	56.1	73.8
A work colleague	36.2	35.0	33.8	40.8
Other	5.3	5.0	5.7	4.9
Know someone diagnosed in last 2 years (%)
Yes	80.6	76.3	79.0	86.4
No	18.5	22.5	20.4	12.6
Missing	0.9	1.3	0.6	1.0

Note. EV phases are based on duration of service of 1 year which in this context is equivalent to the first EV experience (Novice); 2 to 4 years (Transition); and 5 to 6 years (Sustained), respectively. EV = episodic volunteers.

Factor Analysis of the Motives for Volunteering Scale

The 28 items measuring motives for volunteering for the event were analyzed using factor analysis with PCA extraction and orthogonal (varimax) rotation. This analysis showed a six-component solution with cross-loadings above 0.4 for four items (“I want to support the success of the RFL event,” “This event provides great camaraderie within my family/friends,” “I want to raise awareness of cancer,” and “I want to be part of the fight against cancer”) and one freestanding item (“Someone asked me to join the event”). Therefore, these five items were deleted (Costello & Osborne, 2005) and a second PCA with varimax rotation was run. This second analysis revealed a clear four-component structure determined by eigenvalues above 1 and the scree plot which showed four data points above the “break” in the line. All item loadings were above 0.32 (Tabachnick & Fidell, 2001). Component loadings are displayed in Table 2 (bold) including those items with low cross-loadings (below 0.4). Factor analysis using alternative extraction (e.g., principal axis factoring) and rotation (e.g., oblimin) techniques revealed an identical four-factor solution, and thus the PCA results are reported.

Table 2.

Component Loadings for the Motives for Volunteering Scale.

	Component loading
Motives item and component	1	2	3	4
Socializing/enjoyment (25.68% variance)
1. It allows me to spend quality time with family members and/or friends	0.63
2. My friends and/or family encouraged me to join the event	0.57
3. I enjoy the entertainment provided by Cancer Council	0.60
4. I want to have fun	0.79
5. I enjoy walking or running	0.80
6. I feel it keeps me healthy	0.83
7. I want to be with friends	0.86
8. Social gatherings and friendships are an important part of the event	0.79
9. I want to form a group to accomplish a goal	0.69
10. I want to be part of a team challenge	0.68
Celebrating/remembering/fighting back (19.15% variance)
11. I want to celebrate cancer survivors		0.80
12. I want to remember family and/or friends affected by cancer		0.79
13. I want to give hope to people facing cancer		0.87
14. I want to support cancer patients and their families		0.84
15. It is a worthwhile program		0.75
16. I want to support the cause (fight against cancer)		0.76
Benefits (11.81% variance)
17. I want to be recognized at this event that honors my team	0.32		0.71
18. I want to receive a trophy or certificate that acknowledges my team			0.86
19. I want to get tax advantages and tax deduction			0.78
20. I want to get a T-shirt and specialty items			0.83
Financial support (10.25% variance)
21. I want to provide financial support for Cancer Council				0.84
22. I want to raise money for research and programs of Cancer Council		0.38		0.74
23. I want to provide financial support for the RFL event	0.38			0.75

Note. Total variance explained = 66.88 %; bold font represents retained items. RFL = Relay For Life.

Factorability of the data for the second PCA was confirmed with a ratio of participants to item of 12:1, a significant Bartlett’s test of sphericity (approximate χ² = 5082.00, df = 253, p < .001), and Kaiser–Meyer–Olkin measure of sampling adequacy of 0.90 (Costello & Osborne, 2005). The four components reflected motives related to (a) socializing/enjoyment, (b) celebrating/remembering/fighting back against cancer, (c) benefits (e.g., recognition, tax benefits), and (d) providing financial support to the organization, and these explained 66.88% of variance. Cronbach’s alpha coefficients for each component ranged from .83 to .92, item-total correlations all exceeded .5, and inter-term correlations were above .3, thus indicating excellent internal consistency (Field, 2005). Table 3 displays component means, standard deviations, and correlations.

Table 3.

Descriptive Statistics for Episodic Volunteers on Predictor and Outcome Variables.

	1	2	3	4	5	6	7	8	9	10	11	12
1. Gender^a	—
2. Age	−.13*	—
3. Country of birth^a	.01	−.19**	—
4. Socializing/enjoyment motive	.06	−.02	.00	—
5. Celebrating/remembering/fighting back motive	.04	.06	.04	.44***	—
6. Benefits motive	.15**	−.11	.04	.37***	.03	—
7. Financial support motive	.01	.08	.01	.51***	.50***	.11	—
8. PSOC	.06	.14**	−.00	.14**	.17**	−.06	.18**	—
9. Social norm	−.04	.17**	−.00	.37***	.39***	.21***	.39***	.11	—
10. Satisfaction	−.04	.03	−.01	.54***	.57***	.14**	.59***	.14*	.52***	—
11. Commitment	−.02	.01	.01	.60***	.46***	.27***	.65***	.12*	.54***	.65***	—
12. Volunteer retention	−.07	−.01	.07	.52***	.51***	.17**	.62***	.10	.55***	.77***	.71***	—
M	—	42.93	—	4.02	4.59	4.16	2.66	3.62	3.24	4.38	3.76	4.27
SD	—	13.72	—	0.62	0.44	0.65	0.90	0.49	0.64	0.55	0.66	0.61

Note. PSOC = psychological sense of community.

Spearman’s Rho correlations with dichotomous variables.

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

Descriptive Analysis of Predictor and Outcome Variables

Handy et al. (2006) found significant differences in EV categories based on gender, age, and country of birth. As an initial step, we examined correlations between these sample characteristics, and the predictor and outcome variables (Table 3). These characteristics were not correlated with volunteer retention and their inclusion in the regressions predicting retention did not significantly increase the explained variance or change the pattern of results described. As such, the regressions reported in this article do not include sample characteristics. The means, standard deviations, and correlations for predictors (motives, psychological sense of community, social norm, satisfaction, and commitment) and the outcome variable of EV retention (intention to continue volunteering) are also presented in Table 3. Correlations between predictor variables did not exceed .65. Overall, the strongest correlates of EV retention were satisfaction, commitment, and financial support motive, in the same order.

Multiple Regression Analyses Predicting Volunteer Retention

Three multiple regression models assessed the predictors of EV retention for each EV phase of novice (first experience/1 year), transition (2-4 years), and sustained (5-6 years consecutively) and are presented in Table 4. Higher social/enjoyment motives, less benefit motives, higher social norms, and greater satisfaction explained 81% of the variance in Novice EV retention. Greater satisfaction and commitment were associated with increased retention for Transition EVs and accounted for 64% of variance in retention. Increased financial support motives, social norm, satisfaction, and commitment were associated with increased retention and explained 77% of variance for Sustained EVs.

Table 4.

Hierarchical Multiple Regression Analysis for Variables Predicting Novice, Transition and Sustained Episodic Volunteer Retention.

	Novice EV retention			Transition EV retention			Sustained EV retention
Variable	R² (adjusted R²)	F	β	R² (adjusted R²)	F	β	R² (adjusted R²)	F	β
Socializing/enjoyment motive	.81 (.79)	36.18***	.17 *	.64 (.62)	31.82***	−.13	.77 (.75)	37.28***	−.05
Celebrating/remembering/fighting back motive			.02			−.02			−.03
Benefits motive			−.15 *			.07			−.01
Financial support motive			.04			.13			.31 ***
PSOC			−.00			−.06			.06
Social norm			.20 **			.05			.18 **
Satisfaction			.56 ***			.47 ***			.41 ***
Commitment			.12			.38 ***			.19 *

Note. EV phases are based on duration of service of 1 year which in this context is equivalent to the first EV experience (Novice), 2 to 4 years (Transition), and 5 to 6 years (Sustained), respectively. Significant standardized beta weights are in bold font. EV = episodic volunteer; PSOC = psychological sense of community.

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

Discussion

The present study demonstrates the efficacy of applying an integrated theoretical approach to enable a better understanding of the determinants of retention across the episodic volunteering life cycle. In doing so, three phases of volunteering were confirmed: novice, transition, and sustained, each with a specific pattern of antecedents and experiences predicting intention to continue volunteering in the future. Hence, a dynamic model of volunteering behavior that allows for the evolution and development over time of different volunteer typologies with distinct characteristics was supported.

Novice EVs

As expected, EV’s decisions to continue volunteering after their first experience were related to their motives, social norm, and satisfaction. In this novice phase, self-oriented motives such as spending time with family/friends and having fun were linked to EV retention. Other-oriented motives such as fighting back against cancer did not predict EV retention in any phase and this is likely due to overall highly positive endorsement of these motives leaving little variability in responses. In contrast, Handy et al. (2006) found “long-term committed” volunteers and “habitual” EVs at summer festivals had more other-oriented motives than “genuine” EVs, whereas self-oriented motives did not differ across EV groups. One explanation for these diverging findings is the unique episodic volunteering setting in each study, suggesting the importance of conducting future research in multiple contexts to build an evidence base on the contribution of motives to EV retention.

Novice EVs who were more concerned about the expectations of others regarding their volunteering reported increased retention. This result is consistent with traditional volunteer research demonstrating that new volunteers are often exposed to social pressure prior to their decision to volunteer (Penner, 2002), and the perceived expectations of others predict intentions to volunteer in future (Hyde & Knowles, 2013) as well as actual length of service (Finkelstein et al., 2005). Moreover, this finding concurs with anecdotal evidence that many EVs who volunteer for events for the first time are asked by a friend, family member, or colleague to do so. With regard to the impact of experiences on EV retention, results of the present study support the core proposition of the Three-Stage Model of Volunteer’s Duration of Service that satisfaction, rather than commitment, is critical for volunteer retention in the early stages of volunteering (Chacón et al., 2007).

Transition and Sustained EVs

For EVs in the transition and sustained phases, organizational commitment predicted retention although contrary to expectations psychological sense of community did not. Thus, in accord with prior research (Chacón et al., 2007), a commitment to the organization may form over time as EVs continue to volunteer; however, key contributing factors and optimal intervention points to support the development of an organizational commitment have yet to be identified. The lack of predictive capacity of psychological sense of community in the present study may be explained by measurement limitations with the short four-item measure demonstrating low internal consistency. Alternatively, the measure reflected a sense of connection to the community in which EVs reside and therefore may not have fully captured other relevant communities to which EVs feel connected. For instance, EVs in the cancer control setting may feel a sense of connection to the community for which they contribute their time including cancer patients, their families, or RFL events, rather than a community which is defined by geographical boundaries (Omoto & Snyder, 2002). Moreover, Handy et al. (2006) suggest that “habitual” EVs, given the repeat nature of their involvement in events, may be most likely to feel a sense of connection to the event community, compared with “genuine” episodic volunteers. Future testing of the full psychological sense of community measure incorporating multiple salient communities is therefore needed to further clarify the role of psychological sense of community in Novice, Transition, and Sustained EV retention in the current and broader settings.

Unexpectedly, social norm and motives were predictors of sustained EV retention, and satisfaction was a strong contributor to EV retention in both transition and sustained phases. The emergence of social norm as a predictor of sustained EV retention may represent the social nature of the event and long-term connections formed with other EVs at this event. As well, compared with EVs in the novice and transition phases, sustained EVs reported a greater number of close familial connections (e.g., parents, partners, siblings) who had been diagnosed with cancer suggesting that other’s expectations may feature more prominently in their decisions to continue volunteering. Motives were not expected to inform sustained EV retention; however, providing financial support to the organization or the event emerged as a strong motive predicting retention. This motive may represent an additional indicator of sustained EVs’ overall commitment to the organization. In contrast to prior work (Chacón et al., 2007), satisfaction was a significant predictor of retention regardless of EV phase. Thus, once an organizational commitment develops, the importance of satisfaction as a predictor of retention does not diminish but continues to impact throughout the EV life cycle, at least in the current study context. It is therefore essential to understand the determinants of EV satisfaction overall and to explore whether these determinants differ for EVs in the novice, transition, and sustained phases. For example, the degree to which the episodic volunteering experience fulfills initial motives for volunteering or matches original expectations about volunteering (i.e., motive fulfillment; Davis, Hall, & Meyer, 2003) and characteristics of the setting (e.g., event venue) may be critical to ensure EVs in the novice or transition phase have a satisfying experience. By comparison, perceptions about organizational support and appropriate expenditure of funds raised may be more critical for retention of EVs in the sustained phase.

Theoretical Implications

Based on the results of the current study, we propose an Episodic Volunteer Engagement and Retention (EVER) model for further testing and refinement in this and broader episodic volunteering contexts. Specifically, the EVER model builds on available traditional volunteering theories (i.e., Volunteer Process Model and Three-Stage Model of Volunteer’s Duration of Service) and integrates evidence from prior episodic volunteering research to provide a systematic understanding of retention across the episodic volunteering life cycle. Importantly, the EVER model proposes that the key determinants of EV retention evolve over time and offers guidance as to the critical phases when these changes may occur: after the first episodic volunteering experience (novice), during a 2 to 4 year period of episodic volunteering on an irregular basis (transition), and following 5 to 6 or more years of consecutive, regular episodic volunteering (sustained). The following relationships between antecedents, experiences, and consequences are proposed and also displayed in Figure 1: Motives, social norm, and satisfaction predict Novice EV intentions to continue volunteering; psychological sense of community, satisfaction, and organizational commitment predict Transition EV intentions to continue volunteering; and social norm, psychological sense of community, satisfaction and organizational commitment predict Sustained EV intentions to continue volunteering. In turn, for all phases, intention to continue volunteering serves as a direct causal link to actual EV retention. The EVER model may serve as a guiding framework for future empirical prospective tests to identify pathways to EV retention based on duration of volunteering, and may also prove efficacious as a basis to inform understanding about the potential link between episodic volunteering and more traditional forms of volunteering.

Figure 1.

Proposed EVER model.

Practical Implications

Strategies to promote satisfaction, social connections, and organizational commitment may be important to encourage evolution of Novice and Transition EVs to the sustained episodic volunteering phase. Anecdotal evidence from free-text comments at the end of the survey suggests that having sufficient information about episodic volunteer roles for the event as well as event characteristics such as the environment (e.g., location, safety, weather), entertainment, or timing may affect Novice EV satisfaction. One strategy may involve development and distribution of a standard audiovisual information resource (e.g., DVD or online) about the event and volunteer roles to first-time EVs to address information needs. Social norm was an important predictor of intentions to continue volunteering for Novice and Sustained EVs. Thus, expanding social connections beyond team members to include other EVs at events may be a useful approach, and potentially this could be achieved via a “buddy” system involving Novice and more experienced EVs. Also, development of a commitment to the organization is likely critical for EV retention in the longer term. Personal contact from the organization such as via follow-up phone calls or other strategies that enhance the organization and volunteer connection may encourage commitment for Novice EVs. The strong motive to support the organization financially coupled with anecdotal evidence suggests that organizational transparency and information regarding expenditure of funds raised may strengthen transition and sustained EV commitment.

Limitations

Limitations of this study include the cross-sectional design, which precludes inferences about causality, the use of a proxy intention measure for actual retention (Chacón et al., 2007), and sample characteristics. First, future prospective and longitudinal work which includes a measure of volunteer retention behavior is needed to verify the proposed episodic volunteering phases and to give a clearer picture of the processes underpinning EV retention over time. Second, the low reliability of the brief four-item Psychological Sense of Community scale indicates this scale requires further testing to identify potential use of the full measure in future research. Third, qualitative research may also be useful to identify salient communities and additional EV-specific motives, and to provide more in-depth information about the transition from irregular to sustained episodic volunteering or more traditional volunteering roles. Fourth, to control for any potential influence of setting or role, the study was conducted in the specific cancer control setting of RFL events with EVs fulfilling the role of Team Captain. Although an excellent response rate was achieved for the field, caution is needed in generalizing these results to the broader EV population. Moreover, males and EVs younger in age were not well-represented in this study. Future work across multiple settings, with EVs in multiple roles using more gender and age-balanced samples is needed.

Conclusion

The present exploratory study represents one of the few empirical studies using a theoretically driven approach to identify the key determinants of EV retention. It demonstrates the importance of considering critical points in the episodic volunteering life cycle such as novice, transition, and sustained episodic volunteering phases during which the antecedents and experiences informing retention may vary. The data provide important insights on which to base an EVER model to underpin future rigorous and systematic work to address the knowledge gap regarding the determinants of EV retention. Future prospective and longitudinal research, however, is needed to further develop, test, and refine the EVER model in multiple episodic volunteering settings and to assist NPOs in responding to this crucial and increasing phenomenon.

Footnotes

We thank Natalie Wust and Kate White for their invaluable assistance with data collection for this project.

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.

Author Biographies

Melissa K. Hyde,PhD,is a research fellow in psycho-oncology and community engagement at the Griffith Health Institute,Griffith University and Cancer Council Queensland. Her research focuses on understanding and increasing people’s engagement and retention for prosocial behaviors in health contexts.

Jeff Dunn,PhD,is the Chief Executive Officer of Cancer Council Queensland and holds a Professorial appointment with Griffith Health Institute in the School of Medicine. He is actively involved in research on supportive care for cancer patients and has a long-standing focus on community engagement for cancer control initiatives.

Caitlin Bax,BPsychSc,was an honors student in the School of Applied Psychology at Griffith University. This research comprised part of her honors thesis.

Suzanne K. Chambers,PhD,is a health psychologist and Professor of Preventative Health in the Griffith Health Institute at Griffith University. She has more than 20 years of experience in psycho-oncology research and this includes a long-term research focus on the role of volunteering in supportive care for cancer patients and their partners.

References

Beder

Fast

(2008). Episodic volunteering: Why people walk/run for charity. The International Journal of Volunteer Administration, 25(3), 3-13.

Bryen

L. M.

Madden

K. M.

(2006, April). Bounce-back of episodic volunteers: What makes episodic volunteers return? (Working Paper No. CPNS 32). Brisbane, Australia, Queensland University of Technology.

Callero

P. L.

(1985). Role-identity salience. Social Psychology Quarterly, 48, 203-215.

Cancer Council. (2014). History of Relay. Retrieved from http://www.relayforlife.org.au/learn/history/

Chacón

Vecina

M. L.

Dávila

M. C.

(2007). The three-stage model of volunteers’ duration of service. Social Behavior and Personality, 35, 627-642.

Clary

E. G.

Snyder

Ridge

R. D.

Copeland

Stukas

A. A.

Haugen

Miene

(1998). Understanding and assessing the motivations of volunteers: A functional approach. Journal of Personality and Social Psychology, 74, 1516-1530.

Cnaan

R. A.

Handy

(2005). Towards understanding episodic volunteering. Vrijwillige Inzet Onderzocht, 2(1), 29-35.

Costa

C. A.

Chalip

Green

Simes

(2006). Reconsidering the role of training in event volunteers’ satisfaction. Sport Management Review, 9, 165-182.

Costello

A. B.

Osborne

J. W.

(2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10, Article 7. Retrieved from http://pareonline.net/pdf/v10n7.pdf

10.

Davis

M. H.

Hall

J. A.

Meyer

(2003). The first year: Influences on the satisfaction, involvement and persistence of new community volunteers. Personality and Social Psychology Bulletin, 29, 248-260.

11.

Field

(2005). Discovering statistics using SPSS (2nd ed.). London, England: Sage.

12.

Finkelstein

M. A.

Penner

L. A.

Brannick

M. T.

(2005). Motive, role identity, and prosocial personality as predictors of volunteer activity. Social Behavior and Personality, 33, 403-418.

13.

Garner

J. T.

Garner

L. T.

(2011). Volunteering an opinion: Organizational voice and volunteer retention in nonprofit organizations. Nonprofit and Voluntary Sector Quarterly, 40, 813-828.

14.

Handy

Brodeur

Cnaan

R. A.

(2006). Summer on the island: Episodic volunteering. Voluntary Action, 7(3), 31-46.

15.

Hustinx

(2005). Weakening organizational ties? A classification of styles of volunteering in the Flemish Red Cross. Social Service Review, 79, 624-652.

16.

Hustinx

Haski-Leventhal

Handy

(2008). One of a kind? Comparing episodic and regular volunteers at the Philadelphia Ronald McDonald House. International Journal of Volunteer Administration, 25(3), 50-66.

17.

Hyde

M. K.

Dunn

Scuffham

P. A.

Chambers

S. K.

(2014). A systematic review of episodic volunteering in public health and other contexts. BMC Public Health, 14, Article 992.

18.

Hyde

M. K.

Knowles

S. R.

(2013). What predicts Australian university students’ intentions to volunteer their time for community service? Australian Journal of Psychology, 65, 135-145.

19.

Jiménez

M. L. V.

Fuertes

F. C.

Abad

M. J. S.

(2010). Differences and similarities among volunteers who drop out during the first year and volunteers who continue after eight years. The Spanish Journal of Psychology, 13, 343-352.

20.

Macduff

(2004). Episodic volunteering. Walla Walla, WA: Macduff/ Bunt Associates Publishers.

21.

Macduff

(2005). Societal changes and the rise of the episodic volunteer. In Brudney

(Ed.), Emerging areas of volunteering (Vol. 1, No. 2, pp. 49-61). Indianapolis, IN: Association for Research on Nonprofit Organizations and Voluntary Associations.

22.

McCurley

Lynch

(2005). Keeping volunteers: A guide to retention. Olympia, WA: Fat Cat Publications.

23.

Mowday

R. T.

Steers

R. M.

Porter

L. W.

(1979). The measurement of organizational commitment. Journal of Vocational Behavior, 14, 224-247.

24.

Nasar

J. L.

Julian

D. A.

(1995). The psychological sense of community in the neighborhood. Journal of the American Planning Association, 61, 178-184.

25.

Omoto

A. M.

Snyder

(1995). Sustained helping without obligation: Motivation, longevity of service, and perceived attitude change among AIDS volunteers. Journal of Personality and Social Psychology, 68, 671-686.

26.

Omoto

A. M.

Snyder

(2002). Considerations of community: The context and process of volunteerism. American Behavioral Scientist, 45, 846-867.

27.

Penner

L. A.

(2002). Dispositional and organizational influences on sustained volunteerism: An interactionist perspective. Journal of Social Issues, 58, 447-467.

28.

Penner

L. A.

Finkelstein

M. A.

(1998). Dispositional and structural determinants of volunteerism. Journal of Personality and Social Psychology, 74, 525-537.

29.

Productivity Commission. (2010). Contribution of the not-for-profit sector (Research report). Canberra: Australian Government.

30.

Smith

Holmes

Haski-Leventhal

Cnaan

R. A.

Handy

Brudney

J. L.

(2010). Motivations and benefits of student volunteering: Comparing regular, occasional, and non-volunteers in five countries. Canadian Journal of Nonprofit and Social Economy Research, 1(1), 65-81.

31.

Snyder

Omoto

A. M.

(2008). Volunteerism: Social issues perspectives and social policy implications. Social Issues and Policy Review, 2, 1-36.

32.

Tabachnick

B. G.

Fidell

L. S.

(2001). Using multivariate statistics (4th ed.). Boston, MA: Allyn & Bacon.

33.

Vecina

M. L.

Chacón

Sueiro

Barrón

(2012). Volunteer engagement: Does engagement predict the degree of satisfaction among new volunteers and the commitment of those who have been active longer? Applied Psychology, 61, 130-148.

34.

Wilson

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

35.

Won

Park

Turner

B. A.

(2010). Motivations for participating in health related charity sport events. Journal of Venue & Event Management, 1(2), 17-44.