ArticlePDF Available

Volunteer decisions (not) to leave: Reasons to quit versus functional motives to stay


Abstract and Figures

In this article we test whether reasons to quit volunteering can be structured as the commonly used six functional motives to volunteer of Clary et al. (1998). We conjecture that owing to volunteer involvement in an organization, additional contextual factors influence the choice to stop volunteering for that organization. Based on a literature review and a qualitative exploratory analysis, we present items respectively measuring motives to volunteer among active volunteers and reasons to quit among former volunteers in the context of the Scouts and Guides Organization in Flanders (Belgium). We test content-wise symmetry based on expert-rater agreement, while structural symmetry is tested based on factor analyses. Results show that no symmetry can be found. However, additional contextual factors clearly determine the decision to leave an organization. We theorize on how these individual, interpersonal and organizational factors are continuously traded off by volunteers during their involvement in a particular organization.
Content may be subject to copyright.
Human Relations
The online version of this article can be found at:
DOI: 10.1177/0018726712442554
2012 65: 883 originally published online 25 May 2012Human Relations Roland Pepermans
Jurgen Willems, Gert Huybrechts, Marc Jegers, Tim Vantilborgh, Jemima Bidee and
motives to stay
Volunteer decisions (not) to leave: Reasons to quit versus functional
Published by:
On behalf of:
The Tavistock Institute
can be found at:Human RelationsAdditional services and information for Alerts:
What is This?
- May 25, 2012OnlineFirst Version of Record
- Jun 28, 2012Version of Record >>
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
human relations
65(7) 883 –900
© The Author(s) 2012
Reprints and permission: sagepub.
DOI: 10.1177/0018726712442554
human relations
Volunteer decisions (not) to
leave: Reasons to quit versus
functional motives to stay
Jurgen Willems
Vrije Universiteit Brussel, Belgium
Gert Huybrechts
Vrije Universiteit Brussel, Belgium
Marc Jegers
Vrije Universiteit Brussel, Belgium
Tim Vantilborgh
Vrije Universiteit Brussel, Belgium
Jemima Bidee
Vrije Universiteit Brussel, Belgium
Roland Pepermans
Vrije Universiteit Brussel, Belgium
In this article we test whether reasons to quit volunteering can be structured as the
commonly used six functional motives to volunteer of Clary et al. (1998). We conjecture
that owing to volunteer involvement in an organization, additional contextual factors
influence the choice to stop volunteering for that organization. Based on a literature
review and a qualitative exploratory analysis, we present items respectively measuring
motives to volunteer among active volunteers and reasons to quit among former
Corresponding author:
Jurgen Willems, Applied Economics, Vrije Universiteit Brussel, Pleinlaan 2, Brussels 1050, Belgium.
442554HUM65710.1177/0018726712442554Willems et al.Human Relations
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
884 Human Relations 65(7)
volunteers in the context of the Scouts and Guides Organization in Flanders (Belgium).
We test content-wise symmetry based on expert-rater agreement, while structural
symmetry is tested based on factor analyses. Results show that no symmetry can be
found. However, additional contextual factors clearly determine the decision to leave
an organization. We theorize on how these individual, interpersonal and organizational
factors are continuously traded off by volunteers during their involvement in a particular
exploratory and confirmatory factor analysis, inter-rater agreement, reasons to quit,
volunteer dropout, volunteer turnover, volunteer motives
In order to deal with their volunteer force, organizations should have a thorough
understanding of why people want to volunteer and what determines whether or not they
stop after a while (Grube and Piliavin, 2000; Liao-Troth and Dunn, 1999; Machin, 2008).
Farmer and Fedor (2001) confirm that motives to join an organization as a volunteer
differ among people. Therefore, a well developed organizational strategy regarding
attracting, retaining and motivating volunteers ought to be adapted to the particular
objectives of the organization (Callen, 1994; Hartenian, 2007). A major contribution has
been made by Clary and colleagues on the functional motives behind why people
volunteer (Clary and Snyder, 1999; Clary et al., 1996, 1998). From this perspective,
volunteering is seen as behavior that results from different types of motives, or
combinations of them, which can be fulfilled by being involved as a volunteer in a
nonprofit organization (Clary et al., 1996).
However, intuitively the assumption is often made in contemporary literature that, while
fulfillment of functional motives drives people to volunteer, insufficient fulfillment drives
them to quit (the symmetry assumption). Nevertheless, decreasing fulfillment of motives
might not lead directly to lower volunteer satisfaction and/or performance, or to higher
dropout and turnover, as contextual factors might moderate this process (Yanay and Yanay,
2008). A major contribution on this is made by Grube and Piliavin (2000). They discuss the
distinction between the general role identity and the contextual role identity of volunteers.
The general role identity of a volunteer regards those individual characteristics, including
functional motives, which are unique to the volunteer and unrelated to any context. The
contextual role identity deals with those characteristics that a volunteer develops through
continuous interaction with a particular context. As a result, during one’s involvement as a
volunteer in an organization, additional contextual and/or organizational factors might gain
relevance and thus influence the decision whether or not to leave the organization after a
while (Gagné, 2003; Ross et al., 1999; Yan and Tang, 2003).
Consequently, in this article we explore reasons to quit volunteering in contrast to
motives to volunteer in order to assess whether they are conceptually symmetric. If not,
we can infer that other factors than the ones inducing volunteering can determine the
decision to stop volunteering. As a result, our research question is as follows: are reasons
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 885
to quit symmetric to the dimensions of functional motives to volunteer or are additional
contextual aspects relevant? If the latter is true, what are these additional factors?
In the next two subsections we review the functional motives as described by Clary
and colleagues (Clary and Snyder, 1999; Clary et al., 1996, 1998) and the literature on
how contextual factors might become increasingly relevant owing to volunteer
involvement in a particular organization. Subsequently we analyse two related survey
data sets, one from active volunteers and one from former volunteers of the same
organization. Finally, we discuss our findings, summarize our conclusions and postulate
avenues for future research.
Functional motives
Clary and colleagues (Clary and Snyder, 1999; Clary et al., 1996, 1998) distinguish six
functional motives of individuals to participate in volunteer work: ‘values’, ‘understanding’,
‘social’, ‘career’, ‘protective’ and ‘enhancement’. The functional motive ‘values’ deals
with the extent to which one can express personal altruistic values through volunteering.
‘Understanding’ as a functional motive expresses the desire to learn or practice skills and
abilities through volunteering. For the ‘social’ motive, Clary et al. (1998: 1518) describe
its functionality as follows: ‘[v]olunteering may offer opportunities to be with one’s
friends or to engage in an activity viewed favorably by important others’. Within this
definition we distinguish an internal aspect, directed toward other volunteers regarded as
friends, and an external aspect, directed toward important others. We elaborate on this
distinction later on. Based on the ‘career’ motive, volunteering is seen as a way to create
or enhance professional career opportunities. Volunteering as a means to compensate for
own negative feelings, such as guilt or sorrow, is encompassed in the ‘protective’ motive.
In the context of the ‘enhancement’ motive volunteering is a means to enhance self-
esteem, personal satisfaction and/or growth. The extent to which these functional motives
are satisfied by the opportunities available in the volunteering environment is argued to
determine the alignment between an individual and an organization, and the commitment
to (continue to) participate in the organization (Clary and Snyder, 1999). Functional
motives have been studied mainly in relation to other individual characteristics, such as
demographics, volunteer behavior, volunteer activities (Clary et al., 1996), task preferences
(Houle et al., 2005), commitment, citizenship, antisocial behavior, health and socio-
economic achievements (Wilson, 2000). In addition, functional motives have been applied
and adopted for different sub-domains such as for sports volunteers (Hoye et al., 2008;
Wang, 2004) and volunteers in board positions (Inglis and Cleave, 2006).
Contextual factors
Harrison (1995) argues that the time allocation of volunteers to a particular organization
is episodic and complex. They continuously trade off four aspects based on their past
experiences and future expectations: (1) potential return of future volunteer work; (2) the
extent to which people from one’s environment consider volunteer work important;
(3) the opportunities to volunteer; and (4) the clarity of volunteer benefits to the
individual. In the end, this trade-off influences the decision to stay at or to leave an
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
886 Human Relations 65(7)
organization when benefits do not exceed costs anymore. As a result, we distinguish
different factors continuously traded off by volunteers. From an individual perspective,
emotions, perceptions and attitudes of volunteers change owing to their involvement in
an organization, resulting over time in changed behavior, and finally leading to volunteer
retirement (Haski-Leventhal and Bargal, 2008; Hooghe, 2003a). From an interpersonal
perspective, when volunteers feel trusted, helped and appreciated, they will feel more
related to the group of other volunteers and the activities performed collectively (Woolley,
1998). Furthermore, volunteers within groups can influence each other regarding
increasing or decreasing volunteer efforts through socialization and/or signaling
mechanisms (Becker and Connor, 2005; De Cooman et al., 2009; Haski-Leventhal and
Bargal, 2008; Linardi and McConnell, 2011). As a result, the social interaction with other
volunteers seems relevant for the individual decision as to whether or not to stay as a
volunteer in an organization. From an organizational perspective, volunteers evaluate the
extent to which their expectations are met by the organization in order to determine their
level of involvement (Farmer and Fedor, 1999; Richard et al., 2009). Organizations that
want to manage their volunteer turnover and/or dropout, require insights into the contextual
factors influencing volunteer decisions (Miller et al., 1990). By creating a favorable
context taking a volunteer’s generic motives as well as specific preferences into account,
organizations can reduce volunteer dropout and turnover (Jamison, 2003). In contrast, a
lack of volunteer support leads to higher turnover (Cuskelly and Boag, 2001).
To our knowledge, the only study linking motives to volunteer and reasons to quit in
one particular context is the article by McLennan et al. (2008). They analyse reasons for
resignation among volunteer firefighters in Australia and find four clusters of reasons for
resignation related to personal characteristics (age and health concerns) and their
environment (work and family; the organization in which they volunteer; changing
distance from the place of volunteering). In addition, they find a positive impact of
volunteer benefits and the environment in which people volunteer on their satisfaction
and intention to remain with the organization.
In sum, we conjecture that the choice of whether or not to continue volunteering for a
particular organization is determined by a complex set of individual, interpersonal and
organizational factors shaped through the interaction of a volunteer with the organization
(Ross et al., 1999; Yan and Tang, 2003). We investigate in the next section what these
factors are and whether they can be assumed symmetric to the functional motives as
described by Clary and colleagues (Clary and Snyder, 1999; Clary et al., 1996, 1998).
Within the context of a particular volunteer organization – the Scouts and Guides
Movement in Flanders (Belgium) – we have probed among the active volunteers for their
motives to volunteer, while among those who left during the last two years, we probed
for the reasons why they stopped volunteering. Items are generated based on insights
from literature, mainly from Clary and colleagues (Clary and Snyder, 1999; Clary et al.,
1996, 1998), and from an additional qualitative analysis within the case organization
(open question surveys and focus panels. Details are given below.).
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 887
We conducted two expert ratings to assess the extent to which two judges who were
not involved in the item generation would sort items about both motives to volunteer and
reasons to quit into the six dimensions from Clary et al. (1998). The expert judges were
researchers on volunteering and were given the Clary et al. (1998: 1517–18) description
of the dimensions before sorting the items.
After the item sorting procedure, we analysed the factor structure of the two sets of
items using factor analysis in order to determine the degree to which the factor structures
match the six Clary et al. (1998) dimensions.
Organization context and sample
The Scouts and Guides Movement in Flanders is a federated structure in which 561 local
chapters (groups) perform the day-to-day educational, pedagogic and entertaining
activities for children between six and 17 years old (about 60,000 in total). These
activities are supported by volunteers in sub-teams according to the age of the children.
Activities for children take place during the weekends and vacation periods. Volunteer
tasks mainly focus on these activities, but also include preparatory activities, fundraising
activities, social events, training, etc.
In general, volunteers commit themselves to a particular function for a full year
(September–August). Volunteers start at the age of 18, after the last year of being a
member. The average age of the total active volunteer force is 21.4 (median = 20, standard
deviation = 6.0). The official maximum age is 35 for volunteers directly involved with
children. The average age of volunteers quitting the organization in 2009 or 2010 was
22.9 (median = 22, standard deviation = 4.8). Volunteers stay on average about four to
five years, meaning that the yearly turnover is about 20 percent. In December 2010 about
11,000 volunteers were active in the organization. In 2009 and 2010 (August) about 5000
volunteers did not formally renew their commitment.
Online questionnaires were sent to current as well as former volunteers based on their
personal email addresses, which were available at the central registration database of the
organization (about 90% availability). About one week after the initial invitation email,
reminders were sent to those respondents who had not completed the survey at that time.
From the 9812 active volunteers that we invited, 3034 started the survey, and 2212
completed the survey (22.54%). We invited 4974 former volunteers, from which 1264
started the survey and 1085 completed the survey (21.81%).
Item generation
A multiphase procedure was followed to generate items for motives to volunteer and for
reasons to quit. On the one hand, the items were generated inspired by literature (Clary
and Snyder, 1999; Clary et al., 1996, 1998; Inglis and Cleave, 2006; Tschirhart et al.,
2001; Wang, 2004; Ziemek, 2006). On the other hand, additional qualitative steps were
taken to assure as much exhaustiveness as possible. We consider this a necessary step to
verify whether no other factors might be important within the particular organization of
our study.
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
888 Human Relations 65(7)
An open-question survey was sent to and answered by a sample of 17 active or former
volunteers, probing for motives to volunteer and reasons to quit. In addition, we held two
focus panels, one with a group of active volunteers and one with volunteers that had quit,
in order to find the right level of detail. Subsequently, for both groups of respondents a
separate set of items was developed and pretested to improve exhaustiveness, clarity and
practical relevance. In the end the respective questionnaires included 43 items on motives
to volunteer, and 47 items on reasons to quit. Respondents had to answer for each
statement how applicable it was for them on a seven-point Likert-scale: ‘not at all
applicable to me’ (–3) – ‘totally applicable to me’ (+3). Items were randomly shuffled for
each respondent.
Expert ratings
We use Fleiss’ kappa to report on the expert ratings as it enables us to assess agreement
(or accordance) of two or more different categorical classifications, with the same
number of categories per classification (Fleiss, 1971; Gamer et al., 2010). The kappa-
value can range from 0 to 1, with 1 indicating full agreement, supplemented with a
p-value reporting on the significance of the kappa-value.
Table 1 shows kappa-values for both rater analyses. In contrast to the items on motives
to volunteer, not all items on reasons to quit were classified. For 17 out of 47 items at
least one rater indicated that the content was too different from one of the six categories
provided. We chose not to insist on classifying these items in a second rating in order to
avoid artificial and counterintuitive results. Therefore, Table 1 shows results for reasons
to quit for the 30 items classified by both raters. For both expert ratings an overall kappa-
value is calculated. In addition, for each of the categories provided (i.e. the six
Table 1 Fleiss’s kappa inter-rater agreement among raters for motives to volunteer and
reasons to quit
Motives to volunteer (43 items) Reasons to quit (30 items)a
Overall kappa test
Fleiss’s kappa 0.801*** 0.406***
Category-wise kappa-values
Values 0.919*** 0.515***
Understanding 0.758*** 0.712***
Social 0.756*** 0.615***
Career 1.000*** 0.649***
Protective 1.000*** 0.200
Enhancement 0.488*** 0.000b
*p = 0.1; ** p = 0.05; *** p = 0.01.
aOnly 30 out of 47 items were used for analysis (63.8%). Other items were indicated by at least one of the
raters as not to be classifiable in the six dimensions.
bValues below zero are set at zero, totally no agreement found (Gamer et al., 2010).
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 889
motivational dimensions; Clary et al., 1998), a category-wise kappa is given, reporting
the agreement for each particular category (Gamer et al., 2010).
Both overall kappa-values are significantly different from zero, but substantially more
agreement is found for motives to volunteer than for reasons to quit (respective kappa-
values are 0.801 and 0.406). When looking at agreement regarding motives to volunteer,
(extreme) high agreement is found for the categories ‘career’ (1.000), ‘protective’ (1.000)
and ‘values’ (0.919). Substantial agreement is found for the categories ‘understanding’
(0.758) and ‘social’ (0.756), while moderate agreement is found for the category
‘enhancement’ (0.488). For the 30 items on reasons to quit, substantial agreement could
be found for the dimension ‘understanding’ (0.712), while moderate agreement could be
found for ‘values’ (0.515), ‘social’ (0.615) and ‘career’ (0.649). No significant agreement
could be found for ‘protective’ and ‘enhancement’.
Factor analyses: Motives to volunteer
In order to analyse the motivation items surveyed among active volunteers we use a
combined approach (Fabrigar et al., 1999). We start with an exploratory factor analysis
in order to (1) sort out ambiguous and low quality items, and (2) have a starting structure
for confirmatory factor analysis. We continue with confirmatory factor analysis to
evaluate the optimal number of factors to be specified, given the observed data. In
particular, we look at whether a six-factor structure similar to the motivational dimensions
of Clary et al. (1998) can be found. Furthermore, we make a substantive comparison of
the structure found with the classifications made by the expert-raters.
Based on the exploratory factor analysis with oblique rotation, 32 of the 43 items
remain in the analysis. Seven factors emerge, cumulatively explaining 56.4 percent of
the variance of all items. The maximum correlation between factors does not exceed
0.354. Five out of the seven factors seem to match with the motivational dimensions
found by Clary et al. (1998). These factors are ‘values’, ‘understanding’, ‘career’,
‘protective’ and ‘enhancement’. However, the two other factors both seem to relate to the
sixth motivational dimension (‘social’). The six items of one of these factors probe for
the importance of social aspects, however particularly directed toward other people
volunteering within the organization. We refer to this factor as ‘internal social’. In
contrast, the four items of the other factor also probe for the importance of social aspects,
however directed toward people outside the organization. We label this factor ‘external
social’. The items and factor loadings are presented in Table 2. We compare the remaining
32 items with the two classifications made by the raters with the third classification
based on the factor solution. In order to make comparison possible, we reduce the seven
factors to six categories by grouping the ten items of the internal and external social
dimension as a single social dimension. We obtain an overall Fleiss’ kappa-value of
0.767. Category-wise kappa-values are 0.589 (values), 0.582 (understanding), 0.911
(career), 0.897 (protective), 0.643 (enhancement) and 0.903 (social). All values are
significant at p < 0.05. The overall kappa-value indicates substantial match. A high match
exists for the grouped ‘social’ items, ‘career’ and for ‘protective’. Significant, though
moderate, accordance exists for ‘values’, ‘understanding’ and ‘enhancement’.
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
890 Human Relations 65(7)
Table 2 Factor loadings of exploratory and confirmatory seven-factor solution for motives to
My commitment to the Scouts Movement makes me feel important 0.789 0.914*
My commitment makes me feel needed 0.770 0.783*
My efforts for the Scouts Movement make me feel better about myself 0.640 0.877*
My efforts for the Scouts Movement raise my self-esteem 0.581 0.847*
By being a part of the Scouts Movement I can contribute to important
policy issues in our society
0.733 1.085*
Through my efforts for the Scouts Movement, I can show that I am
concerned with social problems
0.707 1.051*
As a result of my commitment for the Scouts Movement, my voice is
heard in society through an organization that is relevant
0.681 1.138*
Within the Scouts Movement I get a better view on what is important in
our society
0.657 0.951*
I had many chances in the past, and therefore I want to contribute to a
good cause through my efforts in the Scouts Movement
0.491 0.759*
At moments when I am busy with the Scouts Movement, I can forget my
personal problems
0.919 1.337*
When I feel bad, I can forget about it by being busy with affairs of the
Scouts Movement
0.821 1.276*
My efforts for the Scouts Movement help me to solve my personal issues 0.618 0.911*
Internal social
My best friends are also in the Scouts Movement 0.787 0.949*
Being committed to the Scouts Movement for me means especially to
accomplish something together with friends
0.670 0.703*
My friends in the Scouts Movement would regret it if I were no part of it
any more
0.603 0.680*
I certainly feel that I share ideas with my fellow leaders, even on matters
that have nothing to do with the Scouts Movement
0.590 0.658*
With other people from the Scouts Movement I can discuss things that I
cannot discuss with people outside the Scouts Movement
0.569 0.823*
Being a part of the Scouts Movement is a way to make friends 0.493 0.688*
By being active in the Scouts Movement I learn to know people who may
be important for my future job
0.767 1.219*
In the Scouts Movement I get in contact with new people who can assist
me with my studies and/or work situation
0.687 1.043*
I can follow trainings through which my chances to find a good job are
0.684 1.118*
Experience in the Scouts Movement looks good on my resume 0.639 0.744*
Through my commitment, I can get a view of what I want to be doing
0.489 0.915*
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 891
Subsequently, based on confirmatory factor analysis we compare fit-statistics for the
seven-factor solution, where we differentiate between internal and external social items,
and the six-factor solution in which we combine all social items in a single factor. For the
evaluation of the fit-statistics of the models specified, we use indicators based on the
discussion by Hu and Bentler (1999) and Marsh et al. (2004): the Root Mean Square
Error of Approximation (RMSEA) with cut-off < 0.06, the Comparative Fit Index (CFI)
with cut-off > 0.95, and the Standardized Root Mean Square Residual (SRMR) with cut-
off < 0.08. In addition, we look at the Parsimony Normed Fit Index (PNFI), for which we
prefer the highest values when comparing different models for the same dataset.
Fit-statistics regard the congeneric model (Graham, 2006), meaning that we assume
that each item loads on one and only one latent variable (selection based on highest
loading of exploratory factor analysis), no correlations exist among item errors, and
variances of latent variables are all restricted to 1. Furthermore, given the significant
correlations between the factors found in the oblique rotated exploratory factor analysis,
we allow the latent variables to be correlated. Results for the six-factor solution are
RMSEA = 0.0634 @ 0.06; CFI = 0.955 > 0.95; SRMR = 0.0555 < 0.08; PNFI = 0.860.
Results for the seven-factor solution are RMSEA = 0.0506 < 0.06; CFI = 0.968 > 0.95;
External social
People I know from outside the Scouts Movement find it commendable
that I am committed to the Scouts Movement
0.760 0.752*
My friends outside the Scouts Movement regard it as an asset that I am a
part of the Scouts Movement
0.704 0.889*
The people that I appreciate myself, consider it positive that I am
commited to the Scouts Movement
0.669 0.739*
My parents, brothers and/or sisters think it is good that I am committed
to the Scouts Movement
0.594 0.617*
I gain experience about things that I cannot learn anywhere else 0.740 0.674*
Being involved in the Scouts Movement enables me to learn by direct
practical experience
0.630 0.655*
By being active in the Scouts Movement I can teach people something
that they cannot learn elsewhere
0.593 0.526*
Being committed in the Scouts Movement ensures that I get a new
perspective on things
0.525 0.655*
I can explore my own strengths 0.516 0.598*
Exploratory Factor Analysis (EFA): Principal components; oblique rotation; lowest initial eigenvalue = 1.076;
cumulative explained variance = 56.28%.
Only factor loadings greater than 0.400 are shown. No items load on two factors with loading larger than 0.400.
Confirmatory Factor Analysis (CFA): Congeneric model (Graham, 2006);
RMSEA = 0.0506; CFI = 0.968; SRMR = 0.0472; PNFI = 0.860.
*All CFA loadings are significant at p < 0.01.
Table 2 (Continued)
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
892 Human Relations 65(7)
SRMR = 0.0472 < 0.08; PNFI = 0.860. Except for a negligible deviation (0.0034) for the
RMSEA of the six-factor solution, all cut-off values for both confirmatory factor analyses
are within the proposed boundaries. However, with a similar level of parsimony (both
PNFI-values are 0.860), the RMSEA, CFI and SRMR values are more convincing for the
seven-factor model.
Factor analyses: Reasons to quit
As we want to assess whether symmetry exists between motives to volunteer and reasons
to quit, we test whether either a six-factor solution or a seven-factor solution would be
appropriate given the actual answers from former volunteers. We use a combined
approach of exploratory and confirmatory factor analysis, similar to our approach on
motives to volunteer (Fabrigar et al., 1999). In addition, we verify whether content of
reasons to quit matches with motives to volunteer by comparing the optimal factor
structure obtained with the rater classification answers.
Based on the exploratory factor analysis with oblique rotation, we leave out nine
ambiguous or low quality items. A six-factor solution explains 46.6 percent of the total
variance, while a seven-factor model explains 49.5 percent. However, based on the
Kaiser-criterion and the scree-plot analysis a ten-factor solution seems appropriate,
explaining 56.6 percent of the total variance. We compare these three potential factor
solutions, based on the fit-statistics of confirmatory factor analyses. Table 3 shows fit-
statistics for all three models (with 38 remaining items). We compare the congeneric
models (Graham, 2006) and allow latent variables to correlate freely. Only the SRMR-
metric of the ten-factor solution meets the required cut-off value. Nevertheless, the
RMSEA and CFI values for the ten-factor solution are close to the cut-off values
proposed. Distances from the cut-off values are clearly higher for the six- and seven-
factor solutions.
Four factors are stable for all three factor solutions, meaning that all the items for
these factors are grouped in the same way. These factors could be labeled as (1) ‘struggles
with other volunteers in the group’, (2) ‘being too old to volunteer in the Scouts and
Guides Movement’, (3) ‘not enough available time’, and (4) ‘unhappy about own
involvement’. The 22 items of these four stable factors are listed in Table 4. Furthermore,
Table 3 Fit-statistics comparing 6-, 7-, and 10-factor solution for reasons to quit (confirmatory
factor analysis)
Fit-statistics Cut-off values 6-factor model 7-factor model 10-factor model
RMSEA <0.060 0.0807 0.0765 0.0626
CFI >0.950 0.887 0.898 0.932
SRMR <0.080 0.0935 0.0921 0.0699
PNFI 0.805 0.807 0.807
Degrees of freedom 650 644 620
χ² (p) 5202.465 (0.000) 4697.354 (0.000) 3234.380 (0.000)
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 893
Table 4 Factor loadings of exploratory and confirmatory 10-factor solution for reasons to quit
Struggles with other volunteers in the group
I became too annoyed about some fellow leaders 0.806 1.622*
I could not agree with the beliefs that some fellow leaders had 0.737 1.602*
I did not feel fully appreciated for the efforts that I did 0.775 1.469*
There were some decisions made in our group that I really did not
0.710 1.443*
My opinion was taken too less into account by some of the fellow
0.711 1.370*
The other leaders of the group were not able to bring as much
commitment as I did
0.721 1.114*
Being too old to volunteer in the Scouts and Guides Movement
I began to feel too old to be a leader 0.736 1.597*
It was time to leave the floor for younger leaders 0.706 1.467*
In case I decided to remain a leader, I would have had to collaborate
especially with people who are much younger than me
0.692 1.337*
I think it is better that the age difference between leaders and members
is not too large
0.672 0.872*
I did not feel connected any more with the young leaders in our group 0.603 0.960*
I believe it is a bit pathetic to be a leader for too long 0.498 0.755*
Most of my best friends also stopped being a leader 0.469 0.952*
Too little available time
I wanted to make time available for other things in my life 0.802 1.084*
I wanted to spent more time for those things that will be important in
my later life
0.794 1.193*
I wanted to have again some more time for myself 0.635 1.301*
Being a leader requires too much of your free time 0.508 1.054*
Being a leader was difficult to combine with my work and/or studies 0.626 0.713*
Few new opportunities
There were only little new things for me to learn anymore 0.729 1.279*
There were only few new challenges to remaining a leader 0.694 1.284*
I couldn’t bring in much new input to our group anymore 0.637 0.897*
I couldn’t bring in much new input to the Scouts and Guides Movement
in Flanders any more
0.607 0.883*
Lack of higher level support
I received too little guidance from above in order to fulfill my duties as a
leader to the utmost
0.747 0.926*
From time to time we were left without guidance too much 0.667 0.979*
The responsibility that we take as leaders is too big 0.625 0.667*
Unhappy about own involvement
I couldn’t invest as much commitment as my fellow leaders 0.789 1.036*
I repeatedly had to find the motivation for extra activities such as group
councils, parents’ evenings, parties and extra activities among the leaders
0.684 1.303*
I wasn’t able any more to be a leader in a qualitative way, and in that
case it is better to quit
0.520 0.974*
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
894 Human Relations 65(7)
the 16 other items are also listed based on the ten-factor solution. Other dimensions we
refer to as (5) ‘few new opportunities’, (6) ‘lack of higher level support’, (7) ‘values
mismatch with the Scouts and Guides Movement’, (8) ‘keep good memories’, (9) ‘next
year’s team preferences’, and (10) ‘other commitments’.
Given insufficient model fit for the six-factor solutions and the initially low agreement
between the content of the items and the original motivational classification (Clary et al.,
1998), we consider comparing the rater classifications and the six-factor solution based
on a Fleiss’ kappa test irrelevant. Furthermore, it is technically not possible to perform a
Fleiss’ kappa-based comparison of the rater classifications with the more likely ten-
factor solution. However, based on qualitative comparison of the ten-factor solution with
the classifications made by the raters, we find that two out of the three items in the factor
‘values mismatch with the Scouts and Guides Movement’ are rated by both raters as
items of the ‘values’ dimension. Three out of the four items of the factor ‘few new
opportunities’ are classified by both raters as items of the ‘understanding’ dimension. No
other structural resemblances are found for the other eight factors.
Values mismatch with the Scouts and Guides Movement
The values of the Scouts and Guides Movement in Flanders are not
consistent with how I think about certain things
0.882 0.948*
I did no longer agree with the policy conducted by the Scouts and
Guides Movement in Flanders
0.850 0.947*
Keep good memories
I wanted to quit at a moment when things were going very well in order
to keep especially good memories
0.873 1.485*
I just had a great year and the chance was small to have another year
like this
0.864 1.406*
Next year’s team preferences
In the year that followed, I probably could not have been in the same
team with the fellow leaders that I preferred (Team: subgroup of leaders
for particular age group)
0.793 1.149*
In the years that followed, I probably could not have been in the team
that I preferred (Team: subgroup of leaders for particular age group)
0.770 0.876*
Other commitments
I wanted to commit myself to another organization 0.757 0.605*
I wanted to spent more time to other hobbies 0.641 1.950*
Exploratory Factor Analysis (EFA): Principal components; oblique rotation; lowest initial eigenvalue = 0.948;
cumulative explained variance = 63.75%.
Only factor loadings greater than 0.400 are shown. No items load on two factors with loading larger than
Confirmatory Factor Analysis (CFA): Congeneric model (Graham, 2006);
RMSEA = 0.0626; CFI = 0.932; SRMR = 0.0699; PNFI = 0.807.
*All CFA loadings are significant at p < 0.01.
Table 4 (Continued)
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 895
Our research question is whether reasons to quit volunteering are symmetric to the motives
to volunteer – often an inherent assumption made in literature on volunteer motives (Clary
and Snyder, 1999; Clary et al., 1996, 1998; Houle et al., 2005; Wilson, 2000) – or whether
to stay or to quit in a particular context depends on a continuous trade-off of personal
motives and contextual factors (Farmer and Fedor, 1999; Harrison, 1995; Haski-Leventhal
and Bargal, 2008; Miller et al., 1990; Richard et al., 2009). Therefore, we have generated
two sets of items, one on motives to volunteer and one on reasons to quit for the particular
context of one organization. First we have tested whether, from a content point of view,
these items match to the six motivational categories described by Clary et al. (1998). As
to the motive items, we find strong overall and category-wise agreement. Only for the
dimension ‘enhancement’ moderate agreement emerges. In contrast, for the reasons to
quit more than one out of three of the items was considered as too distinct to classify in
one of the six motivational dimensions, and only moderate overall and category-wise
agreement is found for the other items (with no significant agreement for ‘protective’ and
‘enhancement’). Therefore, at least from a content point of view, there are limited grounds
to relate reasons to quit to motives to volunteer.
In addition, we explored answers of respectively 2212 active volunteers and 1085
former volunteers to investigate the factor structure of these items. For the motives to
volunteer, a seven-factor solution emerges. Although the number of factors deviates from
the six-dimensional classification of Clary et al. (1998), we find a convincing match with
previous literature on the motives to volunteer. Five of the factors are in line with the
motivational dimensions. These dimensions are ‘values’, ‘understanding’, ‘career’,
‘protective’ and ‘enhancement’. For the two remaining factors we can assume high
conceptual relatedness with the ‘social’ motive as defined by Clary et al. (1998). This can
be inferred from the high inter-rater agreement on the items in these two factors with the
original description of the social dimension (Clary et al., 1998). In contrast, comparing
results of the confirmatory factor analyses for a six- versus a seven-factor solution, fit-
statistics show that a seven-factor solution is more appropriate. Although it is beyond the
scope of this article, one may wonder whether such a seven-factor solution, with a clear
distinction between internal and external social aspects, is an exception for the particular
case organization of our analysis, or whether such distinction could be relevant in other
contexts too. From a content perspective, Tschirhart et al. (2001) refer to the social
dimension as ‘focuses on one’s use of volunteer service to enhance friendship and positive
regard by others’ (p. 426), mentioning both the internal and external aspects as found in our
analysis. From an analytic perspective, Clary et al. (1998), in their seminal work in which
they elaborately validate the items measuring the six motivational dimensions, find no
congruence validity for the social dimension as measured by their original items (referred
to as ‘study 4’ by Clary et al. [1998: 1523]). The need for a distinction between internal
social and external social aspects might be at the base for this shortcoming. Although our
data contribute to the discussion, further research is required.
Assuming a six- or seven-factor solution for the reasons to quit is not appropriate, given
the low percentages of explained variance and the unsatisfactory fit-statistics of the
confirmatory factor analyses. Fit-statistics for the ten-factor solutions are more convincing,
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
896 Human Relations 65(7)
although not fully within the boundaries of the strict cut-off values that we use based on
Marsh et al. (2004). The ten-factor solution does fit, however, within the more common
but less strict cut-off values proposed by Hu and Bentler (1999). Consequently, we can build
on these results for assessing the asymmetry between the structure of motives to volunteer
and the structure of reasons to quit.
Three out of the ten factors describing reasons to quit relate to the motives to volunteer.
Based on the classifications made by the two raters, ‘few new opportunities’ relates to
‘understanding’, while the factor ‘values mismatch with the Scouts and Guides Movement’
relates to the ‘values’ motive. In addition, we can assume that ‘struggles with other volunteers
in the group’ is in line with the internal social dimensions found for the motivational items
used in this study. In addition, several factors underscore the continuous and episodic nature
of the volunteer decision process. Four dimensions refer to how people take past experiences
into account, as well as future expectations and time spent as a volunteer within an
organization: ‘keep good memories’, ‘next year’s team preferences’, ‘unhappy about own
involvement’, and ‘being too old to volunteer in the Scouts and Guides Movement’. These
factors show that, through interaction in a volunteer environment, changed individual
perceptions and interpersonal interaction influence the individual characteristics determining
the choice to stop volunteering. In contrast, the ‘lack of higher level support’ confirms the
importance of organizational aspects. Also, we would like to point out that two factors
mentioned earlier are organization specific (‘being too old to volunteer in the Scouts and
Guides Movement’ and ‘next year’s team preferences’). The Scouts and Guides Movement
is strongly focused on working in teams, meaning weekly and intensive collaboration with
the same small group of team members. We can assume that in this particular context
volunteers express their future concern regarding the fulfillment of the internal social motive
in the dimension ‘next year’s team preferences’. Furthermore, at least for scouting in
Flanders, a strong involvement of young adolescent volunteers is strived for and publicly
promoted as one of its success factors (e.g. official maximum age for leaders and
collaborators, or recurrently mentioning the young character of the organization in public
communication). As a result, within this particular context, volunteers take the consequences
of their age into account for the decision to continue or to stop volunteering. On the one
hand, we can assume that these consequences relate to what volunteers perceive as being
expected from them by the organization (Farmer and Fedor, 1999). On the other hand,
consequences of age broader than the boundaries of the organization might become relevant
too (e.g. moving, starting a new job, having a family, etc.) (Hooghe, 2003b; Hooghe and
Stolle, 2003; McLennan et al., 2008). Finally, other criteria taken into account for the
decision to leave a particular organization relate to the allocation of the time available: ‘not
enough available time’ and ‘other commitments’. As a result, the involvement in a particular
organization is also traded-off in comparison with other activities. This shows that the
fulfillment of functional motives can be strived for in other and/or multiple organizations at
once, depending on the contextual factors of these different organizations.
Based on these considerations, we argue that both motivational and contextual aspects
are traded off for the decision to stop volunteering for a particular organization (Farmer
and Fedor, 1999; Haski-Leventhal and Bargal, 2008). Within this trade-off past, current
and future volunteer costs and benefits are continuously evaluated (Harrison, 1995).
Furthermore, the aspects taken into account relate to individual, interpersonal and
organizational characteristics. While ‘unhappy about own involvement’ exemplifies
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 897
individual aspects, ‘next year’s team preferences’ and ‘lack of higher level support’
respectively exemplify interpersonal and organizational aspects.
As a result, a strict focus on the motivational dimensions to explain the decision
process of volunteers to leave an organization might be too limited, and thus reasons to
quit cannot be assumed to be symmetric to the generic motives to volunteer.
Conclusions and further research
In this article, we aim at scrutinizing the motives why people volunteer and why they
decide to leave an organization after a while. Focusing on the Scouts and Guides
Movement in Flanders (Belgium) we investigate whether motives to volunteer and
reasons to quit can be assumed to be symmetric. We explore two related data sets, one
from active volunteers and one from volunteers that left the organization. We look at
underlying factor structures, the way items and factors conceptually relate to each other,
and to the six-dimensional classification of Clary et al. (1998).
The commonly used motivational dimensions (Clary et al., 1998) clearly emerge from
the data from active volunteers. However, we suggest splitting the ‘social’ dimension, at
least in this context, into two distinct sub-dimensions: internal versus external. In
addition, when we focus on the decision to leave an organization, we find that a limited
number of motivational dimensions influence such a decision. However, additional
factors that result from the interaction of a volunteer in a volunteer environment also
seem to play an important role. Therefore, we conclude that reasons to quit volunteering
in a particular context are not symmetric to the generic motives to volunteer.
In our analyses, we focus on two aspects: motives behind why individuals volunteer
and reasons to quit volunteering. In that sense we do not tackle all dynamic aspects of the
decision process of volunteers. Further research should elaborate on the individual and
contextual aspects of why people join organizations and on the determinants of the
amount of volunteer effort donated. In this analysis we built on a cross-sectional data set
in two groups of respondents, but the evolution of an individual within an organization
(Haski-Leventhal and Bargal, 2008), and also over different organizations (Hooghe,
2003b), cannot be neglected in further analyses. Longitudinal research designs could
substantially support such analyses.
The authors acknowledge the helpful suggestions made by Marlene Walk (University of
Pennsylvania, USA).
This research received no specific grant from any funding agency in the public, commercial, or
not-for-profit sectors.
Becker BW and Connor PE (2005) Self-selection or socialization of public- and private-sector
managers? A cross-cultural values analysis. Journal of Business Research 58(1): 111–113.
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
898 Human Relations 65(7)
Callen JL (1994) Money donations, volunteering and organizational efficiency. The Journal of
Productivity Analysis 5(3): 215–228.
Clary GE and Snyder M (1999) The motivations to volunteer: Theoretical and practical
considerations. Current Directions in Psychological Science 8(5): 156–159.
Clary GE, Snyder M, Ridge RD, Copeland J, Stukas AA, Haugen J and Miene P (1998)
Understanding and assessing the motivations of volunteers: A functional approach. Journal of
Personality and Social Psychology 74(6): 1516–1530.
Clary GE, Snyder M. and Stukas AA (1996) Volunteers’ motivations: Findings from a national
survey. Nonprofit and Voluntary Sector Quarterly 25(4): 485–505.
Cuskelly G and Boag A (2001) Organisational commitment as a predictor of committee member
turnover among volunteers sport administrators: Results of a time-lagged study. Sport
Management Review 4(1): 65–86.
De Cooman R, De Gieter S, Pepermans R, Hermans S, Du Bois C, Caers R and Jegers M (2009)
Person-organization fit: Testing socialization and attraction-selection-attrition hypotheses.
Journal of Vocational Behavior 74(1): 102–107.
Fabrigar LR, Wegener DT, MacCallum RC and Strahan EJ (1999) Evaluating the use of exploratory
factor analysis in psychological research. Psychological Methods 4(3): 272–299.
Farmer SM and Fedor DB (1999) Volunteer participation and withdrawal: A psychological contract
perspective on the role of expectations and organizational support. Nonprofit Management and
Leadership 9(4): 349–367.
Farmer SM and Fedor DB (2001) Changing the focus on volunteering: An investigation of
volunteers’ multiple contributions to charitable organizations. Journal of Management 27(2):
Fleiss JL (1971) Measuring nominal scale agreement among many raters. Psychological Bulletin
76(5): 378–383.
Gagné M (2003) The role of autonomy support and autonomy orientation in prosocial behavior
engagement. Motivation and Emotion 27(3): 199–223.
Gamer M, Lemon J, Fellows I and Singh P (2010) Various coefficients on interrater reliability and
agreement (R package). R Package ‘irr’, version 0.83, available at:
Graham JM (2006) Congeneric and (essentially) tau-equivalent estimates of score reliability: What
they are and how to use them. Educational and Psychological Measurement 66(6): 930–944.
Grube JA and Piliavin JA (2000) Role identity, organizational experiences, and volunteer
performance. Personality and Social Psychology Bulletin 26(9): 1108–1119.
Harrison DA (1995) Volunteer motivation and attendance decisions: Competitive theory testing
in multiple samples from a homeless shelter. Journal of Applied Psychology 80(3): 371–385.
Hartenian LS (2007) Nonprofit agency dependence on direct service and indirect support
volunteers: An empirical investigation. Nonprofit Management & Leadership 17(3): 319–334.
Haski-Leventhal D and Bargal D (2008) The volunteer stages and transitions model: Organizational
socialization of volunteers. Human Relations 61(1): 67–102.
Hooghe M (2003a) Value congruence and convergence within voluntary associations:
Ethnocentrism in Belgian organizations. Political Behavior 25(2): 151–175.
Hooghe M (2003b) Participation in voluntary associations and value indicators: The effect of current
and previous participation experiences. Nonprofit and Voluntary Sector Quarterly 32(1): 47–69.
Hooghe M and Stolle D (2003) Age matters: Life-cycle and cohort differences in the socialization
effect of voluntary participation. European Political Science 3(2): 49–56.
Houle BJ, Sagarin BJ and Kaplan MF (2005) A functional approach to volunteerism: Do volunteer
motives predict task preferences? Basic and Applied Social Psychology 27(4): 337–334.
Hoye R, Cuskelly G, Taylor T and Darcy S (2008) Volunteer motives and retention in community
sport: A study of Australian rugby clubs. Australian Journal on Volunteering 13(2): 40–48.
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
Willems et al. 899
Hu L and Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling 6(1): 1–55.
Inglis S and Cleave S (2006) A scale to assess board member motivations in nonprofit organizations.
Nonprofit Management & Leadership 17(1): 83–101.
Jamison IB (2003) Turnover and retention among volunteers in human service agencies. Review of
Public Personnel Administration 23(2): 114–132.
Liao-Troth M and Dunn CP (1999) Social constructs and human service: Managerial sensemaking
of volunteer motivation. Voluntas: International Journal of Voluntary and Nonprofit
Organizations 10(4): 345–361.
Linardi S and McConnell MA (2011) No excuses for good behavior: Volunteering and the social
environment. Journal of Public Economics 95(5–6): 445–454.
Machin J (2008) Unraveling the complexities of volunteer management in the Third Sector.
Paper presented at the international conference of the International Society for Third Sector
Research, Barcelona, Spain.
McLennan J, Birch A, Cowlishaw S and Hayes P (2008) I quit! Leadership and satisfaction with the
volunteer role: Resignations and organizational responses. Paper presented at the Australian
Psychological Society Annual Conference, Hobart, Australia.
Marsh HW, Hau K and Wen Z (2004) In search of golden rules: Comment on hypothesis-testing
approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and
Bentler’s (1999) findings. Structural Equation Modeling 11(2): 320–341.
Miller LE, Powell GN and Seltzer J (1990) Determinants of turnover among volunteers. Human
Relations 43(9): 901–917.
Richard OC, McMillan-Capehart A, Bhuian SN and Taylor EC (2009) Antecedents and consequences
of psychological contracts: Does organizational culture really matter? Journal of Business
Research 62(8): 818–825.
Ross MW, Greenfield SA and Bennett L (1999) Predictors of dropout and burnout in AIDS
volunteers: A longitudinal study. AIDS Care 11(6): 723–731.
Tschirhart M, Mesch D, Perry JL, Miller TK and Lee G (2001) Stipendend volunteers: Their goals,
experiences, satisfaction, and likelihood of future service. Nonprofit and Voluntary Sector
Quarterly 30(3): 422–443.
Wang PZ (2004) Assessing motivations for sports volunteers. Advances in Consumer Research
31: 420–425.
Wilson J (2000) Volunteering. Annual Review of Sociology 26(1): 215–240.
Woolley F (1998) Social cohesion and voluntary activity: Making connections. Paper presented
at CSLS Conference on the State of Living Standards and Quality of Life, Ottawa, Ontario.
Yan EC and Tang CS (2003) The role of individual, interpersonal, and organizational factors
in mitigating burnout among elderly Chinese volunteers. International Journal of Geriatric
Psychiatry 18(9): 795–802.
Yanay GV and Yanay N (2008) The decline of motivation? From commitment to dropping out of
volunteering. Nonprofit Management & Leadership 19(1): 65–78.
Ziemek S (2006) Economic analysis of volunteers’ motivations – A cross-country study. Journal
of Socio-Economics 35(3): 532–555.
Jurgen Willems is a doctoral student at the Vrije Universiteit Brussel, Belgium. His PhD focuses on
volunteers, governance and organizational objectives in the nonprofit sector. He holds an MSc in
Applied Economics (University of Ghent, Belgium), in Operations and Technology Management
(University of Ghent), and in Quantitative Analysis in the Social Sciences (Catholic University of
Brussels). He worked as a researcher at the Vlerick Leuven Gent Management School in the
Operations and Technology Management Competence Centre. [Email:]
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
900 Human Relations 65(7)
Gert Huybrechts is a doctoral student at the Department of Applied Economics, Vrije
Universiteit Brussel, Belgium. His research is mainly focused on the economic aspects of
volunteering and non-profit organizations. He holds an MSc in Economics (Vrije Universiteit
Brussel) and in Advanced Studies in Economics (Catholic University Leuven, Belgium).
Marc Jegers holds a doctoral degree in economics from the University of Ghent, Belgium and is
Professor of Managerial Economics at the Vrije Universiteit Brussel, Belgium. His research
interests include nonprofit organizations’ management and governance, financial analysis, and
competition policy. He has published in a wide range of academic journals such as American
Economic Review, Academy of Management Journal, Journal of Health Economics, Corporate
Governance, Journal of Corporate Finance, and most of the academic nonprofit related journals.
He is also the author of Managerial Economics of Nonprofit Organisations (Routledge, 2008).
Tim Vantilborgh is a doctoral student at the Department of Work and Organizational Psychology,
Vrije Universiteit Brussel, Belgium. His research interests centre on psychological contracts,
volunteering, individual differences and the interplay between these individual differences and the
situation. His doctoral dissertation focuses specifically on the psychological contract of volunteers.
His work has been published in journals such as Voluntas: International Journal of Voluntary and
Nonprofit Organizations, Nonprofit and Voluntary Sector Quarterly, and Personnel Review.
Jemima Bidee is a doctoral student in the Department of Work and Organizational Psychology at
the Vrije Universiteit Brussel, Belgium. Her doctoral dissertation focuses on the psychological
processes underlying the motivation to volunteer. Besides research on motivation and volunteering,
her interests include research on individual differences and nonprofit management. Her work has
been published in Voluntas: International Journal of Voluntary and Nonprofit Organizations.
Roland Pepermans is Professor and Head of the Research Unit for Work and Organizational
Psychology at the Vrije Universiteit Brussel, Belgium. He holds a MSc from the Vrije Universiteit
Brussel, from Birmingham University, UK, and a PhD from the Vrije Universiteit Brussel. He
teaches on human resource management, organizational behavior and managerial psychology.
The research unit attempts to build bridges between the profit and nonprofit sectors with
projects on motivation, psychological contracts, reward management, talent management and
political skills. He has authored a wide set of publications in international journals and books.
at Vrije Universiteit Brussel, Bibliotheek on June 29, 2012hum.sagepub.comDownloaded from
... Although volunteers' intention to continue has been widely discussed in the literature (e.g., (Cho et al., 2020;Ferreira et al., 2015;Traeger & Alfes, 2019;Willems et al., 2012), the discussion on how volunteers' intention to continue volunteering under uncertain condition and intense stress and pressure, such as during the COVID-19 pandemic, are scarce. From the practical side, this study is very important due to the significant roles of volunteers in providing support to carry out government programs successfully (Brudney & Kellough, 2000;Brudney & Yoon, 2021). ...
... These burdens and challenges are often perceived by individual volunteers as constraints and hence they hinder the progress of their work. In the worst cases, volunteers will consider these constraints as factors that make them give up their work (Nencini et al., 2016;Willems et al., 2012). ...
... Previous research suggests various internal and external factors to affect the continuing engagement of individuals in their volunteering activities (Cho et al., 2020;Ferreira et al., 2015;Henderson & Sowa, 2018;Nencini et al., 2016;Willems et al., 2012). Internal factors are factors that emerge from an individuals' situation and can be in the form of their motivation, their economic conditions, age, health status, availability of time, and expertise. ...
Full-text available
This study aims to test an integrated model that explains volunteers’ intention to continue volunteering in an uncertain and turbulent environment such as the one caused by the COVID-19 pandemic. We argue that volunteers’ intention to continue is influenced by enthusiasm and personal constraints. Specifically, we assert that enthusiasm is influenced by intrinsic motivation, teamwork climate, and role ambiguity. Further, we propose the antecedents of personal constraints to be role ambiguity and perceived support. We conducted an online survey of potential respondents who were purposively sampled as active volunteers during COVID-19 pandemic. A total of 202 responses were complete and utilized for data analysis using a structural equation model. We found that enthusiasm has a positive influence on intention, while personal constraints has negative influence on intention. Further, we found that intrinsic motivation and teamwork climate positively influences enthusiasm, but enthusiasm is negatively influenced by role ambiguity. Similarly, personal constraints were found to be positively influenced by role ambiguity but negatively influenced by perceived support. The results of this study contribute to the discussion regarding the mechanism that increases volunteers’ intention to continue volunteering. Enthusiasm was found to be one of the factors, and its antecedents were empirically tested. In addition, the dual influences of role ambiguity imply the existence of two different paths to increase volunteers’ intention, which contributes to a deeper understanding of the formation of intention to continue volunteering. These findings provide insights and strategies on how to manage and retain volunteers for government and NPOs.
... The Volunteer Function Inventory (VFI) is devised under the guidance of functionalist theory (Clary et al., 1998). Most studies closely related to the green cluster of keywords choose research variables and organize a research framework based on VFI (Clary and Snyder, 1999;Allison et al., 2002;Finkelstein et al., 2005;Houle et al., 2005;Liao-Troth, 2005;Mowen and Sujan, 2005;Kim et al., 2010;Willems et al., 2012;Bang et al., 2013;Oostlander et al., 2014;Alexander et al., 2015;Khalemsky et al., 2020). Moreover, papers related to the green cluster of keywords combine functionalist theory and role identity theory to analyze the service motivation of volunteers. ...
... Volunteer burnout is also an important keyword in the red cluster. After its proposal (Haski-Leventhal and Bargal, 2008) during the analysis of volunteer service stages of Israeli volunteers in 2009, volunteer burnout has been studied as a factor that influences volunteer motivation and leads to the loss of volunteers (Willems et al., 2012). ...
Full-text available
Volunteers play an indispensable role in several major events and activities. The purpose of this study is to review studies on volunteer motivation from 2000 to 2021 and to discover the development trends in this field. The Web of Science Core Collection is the main literature data resource, from which 162 papers on volunteer motivation published in the SSCI were selected. Using two visualization analysis tools, CiteSpace and VOSviewer, this study conducts bibliometric analysis and systematic review from multiple dimensions, identifying the authors, countries, institutions, and journals with high productivity in this field. Additionally, we explored highly cited papers, authors, and journals in this field. This study aims to find the research hotspots and theoretical basis through co-occurrence analysis and cluster analysis of keywords and explore the evolution through the time zone map drawn with CiteSpace. Moreover, we focus on the influence of Chinese and Western cultures (represented by China and the United States) on volunteer motivation. It was found that Chinese volunteers were more affected by collectivism, whereas American volunteers were more affected by individualism. The conclusion of this study constructs a clear framework for research on volunteer motivation, which provides researchers with a deeper and thorough understanding of the connotation of volunteer motivation, providing guidance and support for future research in this field.
... 246). Similarly, in their survey study of motivations and reasons to quit among Flemish Scouts, Willems et al. (2012) found a range of possible reasons to quit, including practical constraints (e.g. time pressure), group processes (conflicts or lack of appreciation) and factors related to motivation (including a lack of learning, impact or challenge). ...
... 250-251). Volunteers' motivations, and their everyday experiences in a specific group or initiative, both influence the decision to stay or to quit (Willems et al., 2012). ...
Full-text available
Human actions may be responsible for much destruction of nature, but it can also be a key solution. Around the world countless volunteers take action for nature: through citizen science, through activities such as planting trees, by organizing excursions, you name it. But who are these nature volunteers? What is it they do, and why do they do it? These questions are rarely asked. This thesis aims to answer these questions based on several surveys among large numbers of Dutch green volunteers. Contributing to nature conservation and a personal connection to nature were found to be key motivations for the volunteers, regardless of their activities. Other diverse motivations also play a role: these include being outside, learning about nature and working together with others. In terms of profile, the high average age and level of education among the respondents is notable. (Re)connecting humans and nature is urgent and important, how can we stimulate this? Suggestions for nature organisations include further strengthening their support and appreciation of their volunteers. In addition, appealing to diverse motivations can help enthuse a wider public.
... Moreover, not all motives relate to sustained participation or dropout. Willems et al. (2012) argue that only the value, understanding, and social functions are related to factors mentioned as reasons to suspend volunteering. ...
Despite the growing interest in student volunteerism, few students volunteer, and volunteer organisations struggle to retain those who do. We explore motivating factors, expectations, and demotivating factors as they relate to retention in student volunteerism Participants were selected from six volunteer projects associated with three South African universities in the Western Cape region. Seventy active volunteers (N = 70) responded to four open-ended questions, analysed using thematic analysis. Volunteer leadership is central to student volunteers' expectations and demotivating factors. Good leadership fulfilled student volunteers' expectations, but student volunteers were demotivated in the face of poor leadership. Six indicators of good leadership were identified. This included efficiency, regular and effective communication, being organised, passionate, and dedicated along with being supportive and facilitating volunteer growth. We argue for the importance of including contextual factors such as volunteer leadership when designing a student volunteer retention framework.
... The functional model created by Clary et al. is the most widely cited and used in the research literature for assessing volunteer motivations (Agostinho & Paco, 2012;Willems et al., 2012;Wilson, 2012). The VFI model has been adapted to a variety of languages, including Spanish (Chacón & Dávila, 2005), Italian (Marta et al., 2006), Chinese (Wu, Wing Lo & Liu, 2009), German (Oostlander, Guentert, Van Schie, & Wehner, 2014). ...
Full-text available
In following paper, based on Clary et al.’s (1998) Volunteer Functions Inventory model, author wants to investigate which group of factors is the most and least important motive for participation in sports events volunteering in Poland for the study group. Data were collected in cooperation with two leading Academic Sports Federation – AZS Warszawa, AZS Kraków. 87 sports volunteers took part in the online survey. The results show that the Values and Understanding factors attained the highest mean score, whereas the lowest scores were for the Social factor. This result is obtained in both age and sex group. Data show that Career factor was a significantly more important motive for people who want to connect a career in the sports industry. The research did not show any statistically significant relationships between the rank of the event and individual factors.
... Volunteering is advocated as a key way to maintain productive engagement in later life (Carr et al., 2015), and older volunteers typically live longer, healthier, and happier than those who do not volunteer (Anderson et al., 2014;Burr et al., 2021). More recently, studies have begun to assess the dynamics of volunteering (i.e., transition in or out of volunteer activities), examining positive outcomes of initiating a volunteer role and reasons why volunteers quit (Carr et al., 2018a(Carr et al., , 2018bTang et al., 2010;Willems et al., 2012). The common goal of these studies is to retain more older adults as volunteers and maximize the benefits of volunteering to older populations. ...
Objectives: Research has extensively documented the concurrent benefits of being a volunteer (versus a non-volunteer), but little is known about older adults who once served as a volunteer but then stopped at some point in their lives (i.e., former volunteers). The current study tracked changes in older adults’ overall life satisfaction and compared these changes among former volunteers, continuous volunteers, and continuous non-volunteers. We also examined whether self-perceptions of aging may serve as a long-term psychological buffer and protect former volunteers’ life satisfaction after they quit volunteering. Method: Data were from the Health and Retirement Study (2006–2016). A pooled sample of participants age 50+ (N = 10,441) indicated volunteer behaviors every other year, and we identified volunteering dynamics based on their volunteering history across 4 waves (8 years). Participants reported on self-perceptions of aging and life satisfaction in the Leave Behind Questionnaire once every 4 years. Results: Continuous volunteers reported greater subsequent life satisfaction than former volunteers and continuous non-volunteers 4 years later, when we adjusted for their baseline life satisfaction. Yet, the difference between continuous volunteers and former volunteers was absent among participants with more positive self-perceptions of aging. Discussion: This study reveals a potential discontinuity in the benefits of volunteering as older adults transition out of their volunteer activities. Findings, however, also reveal individual differences by self-perceptions of aging, offering suggestive evidence that may refine interventions to prolong the benefits of volunteering.
... Early research suggests that a high percentage of volunteer work is episodic in nature and does not correlate to an initial intention to volunteer (Dunn, Chambers, & Hyde, 2015). In addition, the motivation to volunteer depends on multiple reasons: it can be driven by the search for "a rewarding experience" (Cnaan & Goldberg-Glen, 1991, p. 281) or can be seen as a reflection of self-perceived relationships with others (Clary et al., 1998;Willems et al., 2012). It is worth mentioning, however, that these studies privilege volunteers with economic and educational advantages, and that is why it is important to look at the community members and/or former participants of learning programs, and their rationalisation for volunteering in community-learning programs. ...
Full-text available
Many community-based English language learning programs rely on volunteers to lead classes. While some of these volunteers have some teacher training, the majority are not professional educators. The question of these non-professionals’ understanding of what constitutes facilitation of language learning in adult education context, remains underexplored. This paper presents the findings of a small-scale study conducted within a community-based language learning program with four volunteer facilitators. Volunteer facilitators were interviewed on a range of topics related to their role in the program, peer-to-peer interaction, and the impacts of volunteering in their lives. Analysis of facilitator interviews, with reference to program’s guiding educational principles, reveals the following positive factors of the program: informal nature of the community, flexible design of the program, peer-to-peer interaction, and support from program staff. However, the findings also highlight that facilitators’ perspectives and practices varied significantly, due to their different lived experiences, motives for volunteering and linguistic background. This study highlights promising practices, which could serve to design sustainable community-based English language learning programs for adults.
... Most of the researches on volunteer activities focus on the motivation and resources of volunteers. Besides, more and more studies have shown that the motivation is in a dynamic process (Haski-Leventhal & Bargal, 2008;O'Toole & Grey, 2015;Willems et al., 2012). Volunteers devote passion and effort into volunteer activities to improve and express themselves (Kahn, 1990;Schaufeli & Bakker, 2004;Shantz et al., 2014). ...
Full-text available
Looking at the course of educational development in developing countries, nongovernmental organizations (NGOs), nonprofit organizations (NPOs), and volunteers have played a significant role in increasing educational resources in underdeveloped areas and moving educational resources to a balanced state. This study used a mixed-methods approach to investigate the underlying factors affecting volunteer teaching activities of Chinese college students. Data were collected from multiple resources using a self-developed questionnaire and semi-structured interviews. It was found that volunteer teaching activities of college students in China were affected by five key factors including motivation ambiguity, confused identity, unspecialized service, weak incentive mechanism, and a lack of supervision mechanism. In addition, exploratory factor analysis revealed that skills, activity experience, and volunteer performance were significantly correlated with the college students’ willingness to participate in volunteer teaching. Informed by the data, we proposed some tentative solutions to solve the challenges and dilemmas from the perspective of consciousness training, systematism cultivation, concept guidance, and the establishment of incentive and supervision mechanism.
Moderating content on social media can lead to severe psychological distress. However, little is known about the type, severity, and consequences of distress experienced by volunteer content moderators (VCMs), who do this work voluntarily. We present results from a survey that investigated why Facebook Group and subreddit VCMs quit, and whether reasons for quitting are correlated with psychological distress, demographics, and/or community characteristics. We found that VCMs are likely to experience psychological distress that stems from struggles with other moderators, moderation team leads’ harmful behaviors, and having too little available time, and these experiences of distress relate to their reasons for quitting. While substantial research has focused on making the task of detecting and assessing toxic content easier or less distressing for moderation workers, our study shows that social interventions for VCM workers, for example, to support them in navigating interpersonal conflict with other moderators, may be necessary.
Zusammenfassung Auf der Grundlage von narrativen Interviews mit Personen, die ein Engagement in einem Wohlfahrtsverband abgebrochen haben, wird ein Spannungsfeld aus bindenden und abstoßenden Erfahrungen im bürgerschaftlichen Engagement diskutiert. Das Feld der Wohlfahrtspflege ermöglicht einerseits innigliche zwischenmenschliche Erfahrungen und ist andererseits durch Ökonomisierung geprägt. Anhand dieser Ergebnisse wird eine Heuristik zum Verständnis von Engagementprozessen vorgestellt, die auf feldspezifische Erfahrungen sowie auf Spannungsverhältnisse fokussiert.
Full-text available
Goodness-of-fit (GOF) indexes provide "rules of thumb"—recommended cutoff values for assessing fit in structural equation modeling. Hu and Bentler (1999) proposed a more rigorous approach to evaluating decision rules based on GOF indexes and, on this basis, proposed new and more stringent cutoff values for many indexes. This article discusses potential problems underlying the hypothesis-testing rationale of their research, which is more appropriate to testing statistical significance than evaluating GOF. Many of their misspecified models resulted in a fit that should have been deemed acceptable according to even their new, more demanding criteria. Hence, rejection of these acceptable-misspecified models should have constituted a Type 1 error (incorrect rejection of an "acceptable" model), leading to the seemingly paradoxical results whereby the probability of correctly rejecting misspecified models decreased substantially with increasing N. In contrast to the application of cutoff values to evaluate each solution in isolation, all the GOF indexes were more effective at identifying differences in misspecification based on nested models. Whereas Hu and Bentler (1999) offered cautions about the use of GOF indexes, current practice seems to have incorporated their new guidelines without sufficient attention to the limitations noted by Hu and Bentler (1999).
Full-text available
Annual resignation rates for Australian volunteer-based fire agencies are about 6-10% of total volunteer firefighter memberships. Four studies investigated issues potentially relating volunteer resignations to leadership. 1. On 396 exit survey returns from former volunteers, reasons given for resigning were: Work/Family needs, 51%; Moved from the area, 38%; Age/Health issues, 28%; Dissatisfaction with the volunteer role, 25%. A major contributor to Dissatisfaction was poor brigade leadership. 2. A survey of 514 new volunteers found that higher levels of Volunteer Satisfaction, and thus reported Intention to Remain, were associated strongly with being a member of a well-led, inclusive, and harmonious brigade. 3. A survey of 1,589 volunteers found that about one third reported negatively on aspects of volunteer/paid staff relationships. 4. Interviews with senior career staff who supervised the activities of volunteer brigades indicated: (a) variability in the quality of leadership in brigades; and (b) differences in staff approaches to brigade supervision. Overall, the findings suggested: (i) the importance of leadership as a factor to be addressed in order to minimise volunteer firefighter resignations; and (ii) a need for agencies to review how they approach volunteer leadership issues.
Full-text available
Despite the widespread use of exploratory factor analysis in psychological research, researchers often make questionable decisions when conducting these analyses. This article reviews the major design and analytical decisions that must be made when conducting a factor analysis and notes that each of these decisions has important consequences for the obtained results. Recommendations that have been made in the methodological literature are discussed. Analyses of 3 existing empirical data sets are used to illustrate how questionable decisions in conducting factor analyses can yield problematic results. The article presents a survey of 2 prominent journals that suggests that researchers routinely conduct analyses using such questionable methods. The implications of these practices for psychological research are discussed, and the reasons for current practices are reviewed. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Volunteering is any activity in which time is given freely to benefit another person, group or cause. Volunteering is part of a cluster of helping behaviors, entailing more commitment than spontaneous assistance but narrower in scope than the care provided to family and friends. Although developed somewhat independently, the study of volunteerism and of social activism have much in common. Since data gathering on volunteering from national samples began about a quarter of a century ago, the rate for the United States has been stable or, according to some studies, rising slightly. Theories that explain volunteering by pointing to individual attributes can be grouped into those that emphasize motives or self-understandings on the one hand and those that emphasize rational action and cost-bene tit analysis on the other. Other theories seek to complement this focus on individual level factors by pointing to the role of social resources, specifically social ties and organizational activity, as explanations for volunteering. Support is found for all theories, although many issues remained unresolved. Age, gender and race differences in volunteering can be accounted for, in large part, by pointing to differences in self-understandings, human capital, and social resources. Less attention has been paid to contextual effects on volunteering and, while evidence is mixed, the impact of organizational, community, and regional characteristics on individual decisions to volunteer remains a fruitful held for exploration. Studies of the experience of volunteering have only just begun to plot and explain spells of volunteering over the life course and to examine the causes of volunteer turnover. Examining the premise that volunteering is beneficial for the helper as well as the helped, a number of studies have looked at the impact of volunteering on subjective and objective well-being. Positive effects are found for life-satisfaction, self-esteem, self-rated health, and for educational and occupational achievement, functional ability, and mortality. Studies of youth also suggest that volunteering reduces the Likelihood of engaging in problem behaviors such as school truancy and drug abuse.
Despite the fact that over half of the people in the US volunteer each year, there is little theoretical or empirical understanding of volunteer performance. In response, this study examined executive-level volunteers’ multiple contributions of personal resources to a national health care advocacy organization. We expected higher contributions when demands from volunteer roles do not exceed desired levels of contribution, interaction with other volunteers is higher, role investments are higher, and motives to join are consistent with organization’s mission. Regression analyses supported the relation of contributions to social interaction, role investments, and volunteer motives. Suggestions for enhancing the level of volunteer contributions to the organization are made.
Drawing on recent conceptual models on volunteer motiva-tions developed mainly in social psychology, this study proposes that motivation for sports volunteerism is a multidimensional construct that comprises five distinct components, namely (1) Altruistic Value, (2) Personal Development, (3) Community Con-cern, (4) Ego Enhancement, and (5) Social Adjustment. A 20-item scale measuring motivations for sports volunteerism was devel-oped using survey data from 935 qualified respondents. Results of confirmatory factor analysis via LISREL software provided rea-sonably adequate support for the five-factor dimensionality, reli-ability, convergent validity, and discriminant validity of the mea-surement scale.