close

Вход

Забыли?

вход по аккаунту

?

2317192

код для вставки
Journal of Community & Applied Social Psychology
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/casp.677
Assessing Volunteer Motives: A Comparison of
an Open-ended Probe and Likert Rating Scales
LORA D. ALLISON,1* MORRIS A. OKUN2 and KATHY S. DUTRIDGE2
1
2
St Margaret’s Episcopal School, San Juan Capistrano, California, USA
Arizona State University, Tempe, Arizona, USA
ABSTRACT
The purpose of the present study was to compare the motives of volunteers (career, esteem,
protective, social, understanding, and value) as assessed by an open-ended probe and the Volunteer
Functions Inventory (VFI) which employs a Likert rating scale. One-hundred-and-twenty-nine individuals, who volunteered for an organization that focuses on episodic volunteering in the community,
completed both measures of volunteer motives and reported their frequency of volunteering for the
organization. The Spearman rank correlation between the rankings of the six volunteer motives in the
two distributions was 0.71. The maximum variance shared between the same motive as assessed by
the two methods was 0.14. Frequency of volunteering for the organization was predicted by the value
(positive predictor) and social (negative predictor) VFI scale scores. A post hoc analysis of the data
from the open-ended probe revealed three additional motives for volunteering—enjoyment, religiosity, and team building. Copyright # 2002 John Wiley & Sons, Ltd.
Key words: accounts; assessment; functional approach; motives; volunteering
INTRODUCTION
Many community agencies face a discrepancy between the demands from customers and
the supply of volunteers (Edwards & Watts, 1983). Thus, the recruitment of volunteers is a
major concern for administrators of community agencies (Brudney & Brown, 1990).
Clary, Snyder, and Ridge (1992) have discussed how the identification of the motives
expressed by potential volunteers can contribute to their recruitment. Understanding
why individuals are interested in volunteering enables volunteer co-ordinators to organize
their recruitment efforts around themes that are focused on fulfilling the specific needs of
their target group. To be successful, recruitment campaigns must convince potential volunteers that volunteering, relative to other possible activities, is a viable way to address their
* Correspondence to: Lora D. Allison, Director, Middle School Community Service and Student Activities,
St Margaret’s Episcopal School, 31641 La Novia, San Juan Capistrano, CA 92675, USA.
E-mail: [email protected]
Copyright # 2002 John Wiley & Sons, Ltd.
Accepted 19 February 2002
244
L. D. Allison et al.
motivational concerns. In the present study we compared two different approaches (an
open-ended probe versus a Likert rating scale) for assessing volunteer motives.
Theories of motivation
Psychologists who embrace a functional approach and sociologists who embrace a symbolic approach differ in their views of motivation (Clary & Snyder, 1999; Scott & Lyman,
1968). According to the functional approach, motives represent the functions served by
actions. The same action (e.g. volunteering) can serve different functions. These functions
involve the conscious desires of individuals. In contrast, according to the symbolic
approach, motives represent ‘accounts’ generated to justify or to excuse actions. The symbolic approach focuses on the subjective meanings that individuals attach to behaviour.
Motives involve interpretations and explanations of one’s actions. Although Scott and
Lyman (1968) initially examined the strategies used by people when they are asked to verbalize their motives for antisocial behaviours, Weinstein (1978) observed that individuals
develop accounts for prosocial behaviours as well.
Measures of volunteer motives
The functional approach typically involves assessing motives with Likert rating scales.
Because the symbolic approach focuses on generating accounts for one’s actions, it stands
to reason that motives would be assessed via open-ended probes. Researchers have studied
motives for volunteering using both open-ended probes (Nathanson & Eggleton, 1993) and
Likert rating scales (Clary et al., 1998; Cnaan & Goldberg-Glen, 1991). In a study of
ombudservice in nursing homes, Nathanson and Eggleton (1993) asked respondents
why they volunteered. They coded responses to the open-ended probe using six categories.
In descending order, the motives were advocacy (37.7%), giving back to the community
(28.3%), having had a family member in a nursing home (19.8%), gaining job experience
or fulfilling an educational requirement (6.6%), empathy for elders (4.7%), and no information provided (2.8%). The type of motive mentioned by the volunteer was not related to
whether the volunteer fulfilled his or her agreed upon 1-year term of service.
The Volunteer Functions Inventory (VFI) represents the most comprehensive set of
Likert rating scales for assessing motives for volunteering (Clary et al., 1998). Based on
an analysis of the empirical research, Clary and his associates (Clary et al., 1998) identified
six motives for volunteering. These volunteer motives include: (1) developing and enhancing one’s career (career); (2) enhancing and enriching personal development (esteem); (3)
conforming to the norms of, or establishing norms for, significant others (social); (4)
escaping from negative feelings (protective); (5) learning new skills and practicing underutilized abilities (understanding); (6) expressing values related to altruistic beliefs (value).
The VFI contains five reasons for each of these motives. Respondents are asked to report
how important or accurate each reason for volunteering is for them personally. Responses
are made on a Likert rating scale with end points of one and seven.
Comparison of the means in a middle-aged sample for the six VFI scales reveals that the
motive rated as most important was the value motive (M ¼ 5.82), followed by the understanding motive (M ¼ 4.91), and the esteem motive (M ¼ 4.27) (Clary et al., 1998). On
average, career, protective, and social motives were rated as being less than a moderately
important reason for volunteering. According to the functional approach, volunteer interest and commitment are jointly determined by whether there is a match between the
motives that are most salient for a volunteer and the opportunity structures associated with
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Assessing volunteer motives
245
the volunteer experience. Nevertheless, several researchers have reported an additive
effect of value motivation on frequency of volunteering (Clary & Orenstein, 1991; Okun,
1994; Omoto & Snyder, 1995; Penner & Finkelstein, 1998). In other words, as the level of
value motivation to volunteer increases, individuals volunteer more frequently and exhibit
a greater commitment to continue as volunteers.
Purpose of the study
With the VFI, respondents rate the importance of each listed reason for volunteering. An
alternative approach to assessing volunteer motivation is to use an open-ended probe in
which respondents must generate their own account of why they volunteer. With a Likert
rating scale, respondents presumably are indicating the extent to which volunteering fulfills various conscious needs. With an open-ended probe, respondents presumably are providing an explanation that justifies serving as a volunteer. To the authors’ knowledge,
researchers have not assessed motives for volunteering using both Likert rating scales
and open-ended probes. The purpose of the present study was to compare the motives
of volunteers as assessed by the VFI and by an open-ended probe. We posed three research
questions. First, to what extent is the rank order of motives for volunteering similar across
the two methods of assessment? Second, for each of the six volunteer motives, to what
extent are VFI scale scores correlated with scores derived from the open-ended probe?
Third, which method of assessment of motives for volunteering is associated with better
prediction of frequency of volunteering for an organization?
METHOD
Organization
This study was conducted with volunteers from Make A Difference (MAD), a non-profit
organization, located in Phoenix, Arizona. MAD focuses on recruiting employees for episodic forms of volunteering in the local community. It was designed to have qualities that
are particularly appealing to corporations and corporate employees. These include: strong
project management, schedule flexibility, commitment flexibility, team-based programmes, convenient project times, and projects that meet a diversity of community needs.
Volunteers engage in service projects that are planned in conjunction with pre-screened
community organizations. Volunteers are not asked to perform jobs which involve fundraising, clerical work, or janitorial work.
Population and sample
Because MAD does not maintain a separate database for active and inactive volunteers, it
was necessary to define the population as all individuals that had volunteered at some point
within the organization’s 8-year existence (N ¼ 1548). A sample was drawn from this
population by selecting every other name that appeared on their mailing list (N ¼ 774).
Procedure
The survey used in this study was a self-administered mail questionnaire. Volunteers were
sent the survey through first-class mail. A large number of surveys were returned due to
insufficient address information (N ¼ 128). These unreachable respondents reduced the
potential size of the sample to 646 volunteers.
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
246
L. D. Allison et al.
The initial mailing included a cover letter explaining the purpose of the study, the survey, and a prepaid-postage business reply envelope. The cover letter stated, in part:
‘ . . . The questionnaire has an identification number that corresponds to your name on
the mailing list. This is strictly for mailing purposes and your name will never be connected to the survey’. One week after the initial mailing, a follow-up postcard was sent
to potential respondents who had not returned their survey. Two weeks later, a slightly
revised cover letter, a replacement survey, and another postage-prepaid, business reply
envelope were sent to individuals who had not yet replied. A total of 195 surveys were
received for a response rate of 30.2% (195/646).
Seventy-eight per cent of the respondents were female. Thirty-six per cent of the sample
was married. Eleven per cent classified themselves as ethnic minorities. Two-thirds of the
respondents had completed at least a bachelor’s degree. The mean age was 35.37 years old,
with a standard deviation of 9.72 years.
Survey instrument
The survey instrument was divided into four sections. The first section asked volunteers to
describe their participation with MAD. The second and third sections consisted of an
open-ended question asking respondents to list their motivations for engaging in volunteer
work, followed by the VFI. The open-ended question was administered prior to the VFI
to prevent respondents’ answers from being biased from exposure to the reasons listed
in the VFI. The last part of the survey included items designed to gather demographic
information.
Study variables
VFI scale scores. The VFI consists of 30 reasons for volunteering. The instructions for
completing the VFI were as follows: ‘Using the 7 point scale below, please indicate how
important or accurate each of the following possible reasons for volunteering is for you in
doing volunteer work with the MAD organization. In selecting an answer, be realistic
about your motivations’. Respondents rated each reason on a seven-point Likert scale
ranging from not at all an important/accurate motivator (1) to a very important/accurate
motivator (7). VFI scale scores were formed by averaging the responses to the items that
tap each motive.
The reliability of the VFI scales has been assessed using both internal consistency
estimates and test–retest correlations. Across diverse samples, the internal consistency
reliability has been quite good, with values of coefficient alphas typically above 0.80
(Clary, Snyder, & Ridge, 1992). With 30 days between measurements, Clary and his
colleagues (Clary et al., 1998) reported test–retest correlations for scale scores ranging
from 0.64 to 0.78, indicating satisfactory test–retest reliability. In the present study, the
coefficient alphas for the VFI scales were adequate, ranging from 0.75 (protective) to
0.87 (career).
The construct validity of the VFI has been examined using factor analysis. In exploratory factor analyses of college students and middle-aged volunteers, Clary et al. (1998)
identified six interpretable factors that corresponded to the six motives proposed by Clary
et al. (1992). Confirmatory factor analyses of the VFI data indicated that the best fitting
model was the six factor model (Clary et al., 1998). Okun, Barr, and Herzog (1998) tested
three models of volunteer motivation using VFI data collected on two samples of older
volunteers. They found that the six factor model provided a much better fit to the data than
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Assessing volunteer motives
247
either a single factor or a two-factor model. Overall, factor analyses of the VFI provide
support for the construct validity of the VFI.
The criterion validity of the VFI has been examined by testing the prediction derived
from the functional approach that volunteer interest is highest when there is a match
between the motive that is most salient for a volunteer and the theme emphasized in a persuasive appeal for volunteers. Consistent with this prediction, Clary et al. (1998) demonstrated that students’ ratings of the effectiveness of a thematic appeal to volunteer was a
function of the degree to which participants in the study were motivated by the theme
emphasized in the appeal.
Open-ended probe of volunteer motivation. Respondents were asked to list their
motivation(s) for engaging in volunteer work. They were told to be as specific as possible.
Whereas all respondents completed the VFI scales (N ¼ 195), 66% (N ¼ 129) of the
respondents provided the requested information on the open-ended probe of volunteer
motives.
Coding of responses to open-ended probe. For each participant, his or her written
response was transcribed verbatim onto a spreadsheet along with his or her study
identification number. Because our interest was in comparing the responses from the openended probe with the scores from the VFI scales, we coded the responses from the openended probe for the six motives for volunteering assessed on the VFI. To guide the coding
of the responses to the open-ended probe, we provided coders with written descriptions of
each motive (Clary et al., 1992, pp. 6–8; Clary et al., 1998, pp. 1517–1518) and the VFI
items used to assess each motive. The coders (i.e. the second and third authors) were kept
blind with respect to the VFI scale scores of the participants. The coders independently
coded each of the 129 written responses. Each response was evaluated for whether the
respondent described him/herself as being motivated to volunteer by career, esteem,
protective, social, understanding, and value motives.
To determine intercoder reliability, we computed Cohen’s Kappa. Values for Cohen’s
Kappa from 0.60 to 0.74 indicate good intercoder reliability and values of 0.75 or higher
indicate excellent intercoder reliability (Fleiss, 1981). In the present study, the values of
Cohen’s Kappa ranged from 0.68 to 0.82. Disagreements between the two coders were
resolved by discussion. Respondents were assigned values of 0 (motive not mentioned)
or 1 (motive mentioned) for each of the motives.
Ahead, we provide quotes from respondents and indicate the motives assigned to these
quotes by the coders. ‘Volunteering is a self-esteem boost’ (esteem motive). ‘Give back to
the community, improve the community, influence the lives of those less fortunate’
(value motive). ‘Volunteering allows me to escape from my troubles’ (protective motive).
‘Volunteering was a great learning experience for me, it opened up a whole new perspective’ (understanding motive). ‘My motivation for volunteering is my brother and sister. I
want them to see me as a person that wants to help’ (social motive). ‘Since I was interested
in the medical field I felt I needed some hands-on experience for the future’ (career
motive).
Frequency of volunteering for MAD. Respondents were asked to indicate how often
they participated in MAD volunteer projects by selecting one of seven response options.
Response options with coded values in parentheses were: annually (1), semi-annually (2),
quarterly (3), bi-monthly (4), monthly (5), bi-weekly (6), and weekly (7).
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
248
L. D. Allison et al.
RESULTS
Preliminary analysis
We regressed whether respondents completed the open-ended probe of volunteer motivation on the VFI scale scores. After the six VFI scale scores were entered into the logistic
regression model, they were considered for removal. The only variable that was selected
for inclusion in the model was the value motive. The Chi square associated with the value
motive was significant, 2(1, N ¼ 195) ¼ 13.37, p < 0.001. The logit coefficient for the
value motive was 0.61 and the y-intercept was 2.81. Compared to respondents who
did not complete the open-ended probe (M ¼ 5.50), respondents who completed the
open-ended probe had higher value motive scores (M ¼ 6.10).1
Salience of the motives for volunteering
As can be seen in Table 1, inspection of the means for the VFI scales indicates that the
most salient motive was value, followed by understanding and esteem. The means for
the protective, social, and career motives were all below the midpoint of the Likert rating
scale (3.5).
In response to the open-ended probe, participants mentioned the esteem motive most
frequently, followed closely by the value motive. The social, understanding and protective
motives were mentioned by 19 to 28% of the respondents. The career motive for volunteering was rarely mentioned by the respondents.
To compare the salience of the motives across the two methods of assessment, we converted the motive scores to ranks. The Spearman rank correlation between the rankings of
the six volunteer motives in the two distributions was 0.71, which is quite high.
Correlations among motives for volunteering
Table 2 presents measures of association for the relations among the 12 motive scores.
Table 2 includes four types of relations: (a) point-biserial correlations between the
same motives assessed by different measures; (b) point-biserial correlations among
different motives assessed by different measures; (c) Pearson correlations between
Table 1.
Descriptive statistics for motives for volunteering measures (N ¼ 129)
Motive
Measure
VFI
Career
Esteem
Protective
Social
Understanding
Value
Open-ended probe
M
SD
Rank
Percentage
Rank
2.09
4.37
3.03
2.64
4.76
6.10
1.33
1.31
1.29
1.37
1.34
0.84
6
3
4
5
2
1
2
71
28
19
22
67
6
1
3
5
4
2
1
All subsequent analyses were based upon the 129 respondents who completed both the VFI and the open-ended
probe measure of volunteer motivation.
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Assessing volunteer motives
Table 2.
Measures of association among motive variables (N ¼ 129)
1
CARVFI
ESTVFI
PROVFI
SOCVFI
UNDVFI
VALVFI
CAROEP
ESTOEP
PROOEP
SOCOEP
UNDOEP
VALOEP
249
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
2
3
0.47*** 0.37***
0.64***
4
5
0.26**
0.22*
0.09
0.43***
0.52***
0.38***
0.13
6
7
8
9
0.00
0.38*** 0.14
0.03
0.24
0.14
0.29*** 0.07
0.09
0.04
0.13
0.16
0.03
0.00
0.10
0.09
0.39*** 0.10
0.07
0.03
0.07
0.03
0.08
0.11
0.44
0.27
10
11
12
0.09
0.08
0.00
0.15
0.00
0.02
0.17
0.16
0.41
0.24*
0.15
0.08
0.11
0.24**
0.03
0.84*
0.16
0.28
0.08
0.08
0.08
0.04
0.05
0.00
0.25**
0.20
0.13
0.30
0.19
0.06
Note: For the relations among variables 1–6, the measure of association was the Pearson correlation coefficient.
For the relations among variables 7–12, the measure of association was the gamma coefficient. For the relations
between variables 1–6 and 7–12, the measure of association was the point-biserial correlation.
*p < 0.05; **p < 0.01; ***p < 0.001.
CARVFI ¼ Career VFI scale; ESTVFI ¼ Esteem VFI scale; PROVFI ¼ Protective VFI scale; SOCVFI ¼ Social
VFI scale; UNDVFI ¼ Understanding VFI scale; VALVFI ¼ Value VFI scale; CAROEP ¼ Career open-ended
probe; ESTOEP ¼ Esteem open-ended probe; PROOEP ¼ Protective open-ended probe; SOCOEP ¼ Social openended probe; UNDOEP ¼ Understanding open-ended probe; VALOEP ¼ Value open-ended probe.
different motives assessed by the VFI; (d) gamma coefficients between different motives
assessed by the open-ended probes. Of the six point-biserial correlations that involved the
same motives assessed by different measures four were significant ( p < 0.05). The six correlations ranged from 0.15 (for the social motive) to 0.38 (for the career motive). Thus,
when the same motive was assessed, the variance shared between the two measures ranged
from 0.02 (for the social motive) to 0.14 (for the career motive).
In contrast, of the 30 point-biserial correlations that involve different motives assessed
by different measures only one was significant. The point-biserial correlation between the
VFI career motive score and the understanding motive score derived from the open-ended
probe was 0.24. Next, using Chi square tests, we examined the 15 relations among the
motives as measured by the open-ended probe. The only significant relation was between
the career and understanding motives, 2(1, N ¼ 129) ¼ 6.61, p < 0.05. As indicated by the
gamma coefficient of 0.84, volunteers who mentioned the career motive also tended to
mention the understanding motive. In contrast, 10 of the 15 Pearson correlations among
the VFI scales were significant. The Pearson correlation coefficients ranged from 0.00
(career and value) to 0.64 (esteem and protective).
Predicting frequency of volunteering for MAD
First, we inspected the distribution of scores for frequency of volunteering for MAD.
Scores on this variable ranged from 2 (semi-annual volunteering for MAD) to 7 (weekly
volunteering for MAD). The mode, median, and mean were 6 (bi-weekly volunteering for
MAD), 5 (monthly volunteering for MAD), and 4.47 (midway between bi-monthly and
monthly volunteering for MAD), respectively.
Second, we examined the bivariate relations between frequency of volunteering for
MAD and (a) VFI motive scores, and (b) open-ended probe motive scores. Pearson correlation coefficients were used to assess the bivariate relations between VFI motive scores
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
250
L. D. Allison et al.
Table 3.
Predicting fequency of volunteering for MAD (N ¼ 129)
Predictor
Open-ended probe
Career
Esteem
Protective
Social
Understanding
Value
VFI scales
Career
Esteem
Protective
Social
Understanding
Value
Correlation
Standardized partial regression coefficient (beta)
0.05
0.02
0.01
0.03
0.08
0.03
0.00
0.02
0.01
0.03
0.10
0.03
0.12
0.03
0.08
0.16
0.19*
0.23**
0.18
0.17
0.07
0.19*
0.11
0.23*
Note: For the open-ended probe variables, the correlations with frequency of volunteering were computed using
the point-biserial correlation coefficient. For the VFI variables, the correlations with frequency of volunteering
were computed using the Pearson correlation coefficient.
*p < 0.05; **p < 0.01.
and frequency of volunteering, whereas point-biserial correlation coefficients were used to
assess the bivariate relations between motive scores derived from the open-ended probe
and frequency of volunteering. As can be seen in Table 3, frequency of volunteering for
MAD was significantly correlated with VFI understanding scores, r(127) ¼ 0.19, p < 0.05,
and VFI value scores, r(127) ¼ 0.23, p < 0.01.
Next, in separate analyses, we regressed frequency of volunteering for MAD on (a) the
six open-ended probe motive scores, and (b) the six VFI motive scores. The six openended probe motive scores did not account for a significant proportion of the variation
in frequency of volunteering for MAD, F(6, 122) ¼ 0.29, p > 0.05. The value of the multiple R2 was 0.01 and none of the individual standardized partial regression coefficients
(betas) were significant. In contrast, the six VFI motive scores explained a significant proportion of the variation in frequency of volunteering for MAD, F(6, 122) ¼ 2.96, p < 0.01.
The value of the multiple R2 was 0.13. As indicated in Table 3, frequency of volunteering
for MAD was significantly predicted by the VFI value motive scale (beta ¼ 0.23) and by
the VFI social motive scale (beta ¼ 0.19). Thus, frequency of volunteering scores
increase as VFI value scores increase and as VFI social scores decrease.
Post hoc analyses
We carried out two post hoc analyses with the data from the open-ended probe. First, we
wanted to explore whether the number of motives mentioned by volunteers in response to
the open-ended probe would be a better measure of strength of motivation than the mention
of particular motives. Second, we wanted to identify additional motives mentioned by volunteers in response to the open-ended probe above and beyond those assessed by the VFI.
Number of motives and frequency of volunteering. A count was made of the number of
motives mentioned by the volunteers in response to the open-ended probe. Scores ranged
from 1 to 5. The mean and standard deviation were 2.09 and 0.94, respectively. The
Pearson correlation coefficient between number of motives for volunteering mentioned in
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Assessing volunteer motives
251
response to the open-ended probe and frequency of volunteering was 0.05. Thus, neither
the particular motives nor the number of motives mentioned in response to the open-ended
probe were related to frequency of volunteering.
Additional motives for volunteering. The first author coded the 129 responses from the
open-ended probe for additional motives above and beyond those assessed by the VFI.
Three additional motives for volunteering were identified—religiosity, enjoyment, and
team building. An example of the religiosity motive was, ‘Reflections from reading scripture
have reinforced and guided me on this new path’. The volunteer who stated, ‘I pick projects
that are fun’, illustrates the enjoyment motive. The volunteer who mentioned, ‘Working
with co-workers gave us a common goal, to increase morale’, captures the team-building
motive. Seventeen per cent of the respondents mentioned the enjoyment motive, 9%
mentioned the religiosity motive, and 6% mentioned the team-building motive, respectively.
DISCUSSION
The present study was designed to compare two measures for assessing motives for volunteering. Our comparison focused on (a) the correspondence between the rank order of
motives for volunteering with the Likert rating scales and with the open-ended probe,
(b) the shared variance between motives assessed by two different methods, and (c) the
ability of the set of volunteer motive scores associated with each measure to predict frequency of volunteering for an organization.
Participants in the present study, on average, rated the value motive as their most important motive for volunteering, followed by the esteem and understanding motives. The
means for the three remaining VFI motives (career, protective, and social) were all substantially below the midpoint of the scale (3.5). These results parallel those reported by
Clary et al. (1998) for a middle-aged sample. In the present study, when volunteers were
asked to generate their own reasons for volunteering, a somewhat similar picture emerged.
With both the VFI and open-ended probe, the career and social motives were ranked as the
least salient motives for volunteering. The protective motive is ranked in the middle of
both distributions of ranks. Value was the most prominent motive among the VFI scales
and the second most prominent motive among the data generated from the open-ended
probe. The largest discrepancies in ranks occurred for the esteem and understanding
motives. The understanding motive was ranked second among the VFI scales and fourth
among the open-ended probe. The esteem motive was ranked first among the open-ended
probe and third among the VFI scales.
The correlations between the same motive as measured by different measures were quite
modest. Across the six motives, the maximum proportion of shared variance was 0.14. In
part, these modest correlations may be due to the limited variation in the motive scores
derived from the open-ended probe.
Previous research has established that motivation to volunteer predicts frequency of
volunteering (Okun, 1994). In the present study, the ability of the motives for volunteering
to predict frequency of volunteering for an organization differed by type of measure. On
the one hand, the motive scores derived from the open-ended probe were not at all predictive of frequency for volunteering with MAD. On the other hand, the VFI motive scores
explained a significant proportion of the variation in frequency of volunteering for MAD
(R2 ¼ 0.13).
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
252
L. D. Allison et al.
Consistent with previous research (Carlo, Okun, Knight, & De Guzman, 2002), the VFI
value motive was a significant, positive predictor of frequency for volunteering with MAD.
Why should value motivation be positively related to frequency for volunteering with
MAD? One possible explanation is that individuals who hold very strong altruistic beliefs
are more committed to volunteering than individuals who hold moderately strong altruistic
beliefs. A second possible explanation for the positive relation between VFI value motive
scores and frequency for volunteering with MAD is that, as individuals become more
involved in volunteering, they increasingly view volunteering as an opportunity to express
altruistic beliefs.
Interestingly, the VFI social motive was a significant, inverse predictor of frequency for
volunteering with MAD. Why should social motivation be inversely related to frequency
for volunteering with MAD? Perhaps, individuals who are motivated to volunteer by social
pressure from others or to exert influence on others are more interested in the status of
being a volunteer for a community agency than in the role of participating in the volunteer
work of the community agency. In other words, for socially motivated individuals, volunteering on an occasional basis may be sufficient to fulfil their need to be able to claim that
they are an active volunteer of a community agency.
Clearly, motives for volunteering are better predictors of frequency of volunteering
when they are assessed by the VFI as opposed to an open-ended probe. How can this discrepancy be explained? Because we used a single open-ended probe and only coded the
responses for the mention of the various motives, we ended up with a set of dummy variables (i.e. scores of 0 or 1). These dummy variables had relatively small variances (0.02 to
0.22). Perhaps, if we had used a series of open-ended probes and had been able to code the
responses for the intensity of the motives that were mentioned, we would have captured
more variation in the various volunteer motives and variables derived from our open-ended
probe measures of volunteer motivation would have explained more variation in frequency
of volunteering.
Although, as a set, the VFI motive scales were significant predictors of frequency of
volunteering, they accounted for only 13% of the variance. We provide two explanations
for the modest ability of VFI scale scores to predict frequency of volunteering. The first
explanation has to do with the nature of our sample. All of our respondents were active
volunteers, i.e. they volunteered semi-annually or more frequently. Strength of motivation
to volunteer may be more strongly related to the decision to do volunteer work than to the
amount of hours devoted to volunteering.
Our second explanation has to do with the nature of the effects examined in our regression
models. Our regression models included only the main effects of volunteer motives. According to the functional theory of volunteer motivation (Clary et al., 1998), involvement in
volunteer work is a function of the joint effects of individuals’ motives for volunteering
and the opportunities provided in the volunteer work environment to meet their needs.
The correlations among the motive scores differed substantially by method of assessment. In the present study, the median correlation among the different motives measured
by the VFI was 0.26. In previous studies, researchers also have found that VFI motives covary substantially (Omoto & Snyder, 1995). One possibility is that the shared variance
between VFI scale scores is due to common method variance (Anastasi, 1968). For
instance, scores on the VFI may be unduly affected by some irrelevant factor such as
the rating scale. Alternatively, the relations among scale scores on the VFI could be due
to their common dependence on a more abstract and overarching construct reflecting one’s
overall motivation to volunteer (Okun et al., 1998).
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Assessing volunteer motives
253
In contrast, among the motives derived from the open-ended probe, only one of the
relations was significant (career and understanding). The lack of significant relations
between the motives scores derived from the open-ended probe may be due to the marginal
distributions of these variables. For example, in the case of the relation between the career
motive and the esteem motive, the maximum possible value of the Chi square was
only 1.07.
Our post hoc analysis of the data from the open-ended probe revealed three motives for
volunteering not covered by the VFI. To some degree, the religiosity motive and the value
motive from the VFI overlap because they share a concern for the welfare of others. What
distinguishes them is that the religiosity motive entails an explicit affiliation with religious
organizations and materials. None of the VFI motives address volunteering because it is
enjoyable. For some adults, volunteering may be incorporated into their leisure portfolio
because it affords them an opportunity to engage in a pleasurable activity. Finally, team
building emerged as a motive for volunteering among a few of the respondents in the present study. This finding is not surprising because MAD was designed to appeal to corporations and corporate employees.
Implications for practice
An important tool for recruiting volunteers is to assess their motives for volunteering
(Clary et al., 1992). In conjunction with previous research, the results of the present study
support using the VFI to assess volunteer motives. The VFI is easy to administer and to
score. Internal psychometric analyses of the VFI (e.g. internal consistency reliability and
factor analysis) have demonstrated that items ‘behave’ in a way consistent with theoretical
expectations (Clary et al., 1998). External psychometric analyses have shown that volunteer outcomes such as intent to volunteer are a function of the joint effect of VFI motive
scores and the potential needs that can be fulfilled by volunteering (Clary, Snyder, Ridge,
Miene, & Haugen, 1994).
The main drawback of the VFI is that the scores on the six motive scales are substantially correlated with each other. From a functionalist perspective, these correlations make
it difficult to classify volunteers into discrete motive types. Furthermore, if volunteer
recruiters simply use the most salient motive as indicated by the highest VFI scale score
to classify volunteers, then the vast majority of volunteers will be classified as volunteering
in order to fulfil value-related needs. For example, in the present study, we found that for
84% of our respondents, their value VFI scale score equaled or exceeded their other five
VFI scale scores.
One drawback of the open-ended probe based upon the present study is that over onethird of the respondents did not complete it. Perhaps, the rate of missing data would be
lower among potential volunteers, particularly among those that were highly interested
in volunteering. A second drawback of the open-ended probe is that coders must be trained
for several hours before they score the protocols, and scoring the protocols is timeconsuming. A third drawback is the lack of evidence pertaining to whether the open-ended
probe yields a reliable (consistency over time) and valid measure of motives for volunteering. In other studies, researchers have shown that individuals differ in the salience of
volunteer motives as a function of length of service (Gidron, 1984). Yet, in the present
study, the volunteer motive scores derived from the open-ended probe did not predict frequency of volunteering. One advantage of including an open-ended probe is that volunteer
recruiters can identify motives among potential volunteers that are not measured by the
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
254
L. D. Allison et al.
VFI. Therefore, volunteer recruiters may want to use both approaches to assessing the
motives of volunteers.
Limitations of the study
The present study had three limitations. Our response rate was low (30.2%). This low
response rate may be attributable to the inclusion of inactive as well as active volunteers
in the sample. Furthermore, over one-third of the respondents did not complete the openended probe and those that did not had lower mean scores on the VFI value scale than
respondents who did complete the open-ended probe. Also, our sample was recruited from
a single organization that focuses on recruiting employees for episodic forms of volunteering. Given these limitations in the sample, it is interesting to note that our findings with
respect to the open-ended probe and the VFI are consistent with previous research. First, in
both the present study and in previous research (Nathanson & Eggleton, 1993), motives
derived from an open-ended probe were unrelated to level of volunteer participation.
Second, as indicated previously, the order of the means for the highest three VFI scales
(value, understanding, and esteem) was the same in the present study and in the middle-aged sample employed by Clary et al. (1998). Third, the pattern of intercorrelations
among the VFI scale scores is similar to the pattern reported for older adults (Okun et
al., 1998). Fourth, the finding that VFI value scale scores predict frequency of volunteering
replicates the results obtained for college students (Carlo et al., 2002).
A second limitation of the present study is the lack of variability in the distribution
of scores on the open-ended probe. This restriction in variability reduced the likelihood
of finding significant relations between the motive variables derived from the openended probe and frequency of volunteering. Possibly, by including several open-ended
probes, more variability in the distribution of motive for volunteering scores could be
created.
A third limitation of the present study is that the participants were active volunteers during the past year. The generalizability of our findings to potential volunteers is an empirical
question. Finally, both of the measures of volunteer motivation used in the present study
are based upon the assumption that volunteers respond accurately to the types of questions
that are posed (Smith, 1981). Clary et al. (1992) reported that scores on the VFI are unrelated to social desirability. Nevertheless, it is possible that two types of response biases
may influence the motives avowed by individuals for volunteering. On the one hand,
the impression management response bias refers to deliberate effort to make a favourable
impression on an audience. On the other hand, the self-deceptive positivity response bias
refers to the tendency to give self-reports that are honest but overly positive (Paulhus,
1984). By manipulating the assessment context (Paulhus, 1991), researchers could examine the influence of impression management and positive self-deception response biases on
reports of volunteer motivation.
ACKNOWLEDGEMENT
The research on which this manuscript is based was conducted with the support of a grant
from Arizona State University’s Center for Non-profit Leadership and Management.
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Assessing volunteer motives
255
REFERENCES
Anastasi, A. (1968). Psychological testing (3rd ed.). London: MacMillan.
Brudney, J. L., & Brown, M. M. (1990). Training in volunteer administration: assessing the needs
of the field. Journal of Volunteer Administration, 9, 21–28.
Carlo, G., Okun, M. A., Knight, G., & De Guzman, M. R. T. (2002). Prosocial value motivation as a
mediator and moderator of the relations between agreeableness, extraversion, and volunteerism.
submitted for publication.
Clary, E. G., & Orenstein, L. (1991). The amount and effectiveness of help: the relationship of
motives and abilities to helping behavior. Personality and Social Psychology Bulletin, 17, 58–64.
Clary, E. G., & Snyder, M. (1999). The motivations to volunteer: theoretical and practical
considerations. Current Directions in Psychological Science, 8, 156–159.
Clary, E. G., Snyder, M., & Ridge, R. (1992). Volunteers’ motivations: a functional strategy for the
recruitment, placement, and retention of volunteers. Nonprofit Management and Leadership, 2,
333–350.
Clary, E. G., Snyder, M., Ridge, R. D., Copeland, J., Stukas, A. A., Haugen, J., & Miene, P. (1998).
Understanding and assessing the motivations of volunteers: a functional approach. Journal of
Personality and Social Psychology, 74, 1516–1530.
Clary, E. G., Snyder, M., Ridge, R. D., Miene, P. K., & Haugen, J. A. (1994). Matching messages to
motives in persuasion: a functional approach to promoting volunteerism. Journal of Applied
Social Psychology, 24, 1129–1149.
Cnaan, R., & Goldberg-Glen, R. S. (1991). Measuring motivation to volunteer in human services.
Journal of Applied Behavioral Science, 27, 269–284.
Edwards, P. K., & Watts, A. D. (1983). Volunteerism and human service organizations: trends and
prospects. Journal of Applied Social Sciences, 7, 225–245.
Fleiss, J. L. (1981). Statistical methods for rates and proportions (2nd ed.). New York: Wiley.
Gidron, B. (1984). Predictors of retention and turnover among service volunteer workers. Journal
of Social Service Research, 8, 1–16.
Nathanson, I. L., & Eggleton, E. (1993). Motivation versus program effect on length of service:
a study of four cohorts of ombudservice volunteers. Journal of Gerontological Social Work, 19,
95–114.
Okun, M. A. (1994). Relation between motives for organizational volunteering and frequency of
volunteering by elders. Journal of Applied Gerontology, 13, 115–126.
Okun, M. A., Barr, A., & Herzog, A. R. (1998). Motivation to volunteer by older adults: a test of
competing measurement models. Psychology and Aging, 13, 608–621.
Omoto, A. M., & Snyder, M. (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.
Paulhus, D. L. (1984). Two-component models of socially desirable responding. Journal of
Personality and Social Psychology, 46, 598–609.
Paulhus, D. L. (1991). Measurement and control of response bias. In J. P. Robinson, P. R. Shaver, &
L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes: Volume 1
(pp. 17–59). New York: Academic Press.
Penner, L. A., & Finkelstein, M. A. (1998). Dispositional and structural determinants of
volunteerism. Journal of Personality and Social Psychology, 74, 525–537.
Scott, M. B., & Lyman, S. M. (1968). Accounts. American Sociological Review, 33, 46–62.
Smith, D. H. (1981). Altruism, volunteers, and volunteerism. Journal of Voluntary Action Research,
10, 21–36.
Weinstein, R. M. (1978). The avowal of motives for marijuana behavior. The International Journal
of the Addictions, 13, 887–910.
Copyright # 2002 John Wiley & Sons, Ltd.
J. Community Appl. Soc. Psychol., 12: 243–255 (2002)
Документ
Категория
Без категории
Просмотров
2
Размер файла
89 Кб
Теги
1/--страниц
Пожаловаться на содержимое документа