Content uploaded by Marylène Gagné
Author content
All content in this area was uploaded by Marylène Gagné
Content may be subject to copyright.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Motivation and Emotion, Vol. 27, No. 3, September 2003 ( C
°2003)
The Role of Autonomy Support and Autonomy
Orientation in Prosocial Behavior Engagement1
Maryl`ene Gagn´e2,3
Two studies examined individual and environmental forces that affect engagement
in prosocial behavior. Self-determination theory was used to derive a model in
which autonomy orientation and autonomy support predicted satisfaction of three
core psychological needs, which in turn led to engagement in prosocial activities.
In Study 1, college students reported their engagement in various prosocial activi-
ties,andcompletedmeasuresofautonomyorientation,parentalautonomysupport,
and general need satisfaction. In Study 2, volunteer workers completed measures
of autonomy orientation, work autonomy support and need satisfaction at work.
The number of volunteered hours indicated the amount of prosocial engagement.
Results across the studies showed that autonomy orientation was strongly related
to engagement in prosocial behavior, while autonomy support was modestly re-
lated. Need satisfaction partially mediated the effect of autonomy orientation, and
fully mediated the effect of autonomy support. Interestingly, autonomy support
predicted lower volunteer turnover. Implications for how prosocial behavior can
be motivated are discussed.
KEY WORDS: volunteering; prosocial behavior; need satisfaction; self-determination theory;
autonomy support.
1Special thanks to Edward Deci, Richard Ryan, and Craig Barclay for their helpful comments during
all stages of these projects. I also want to thank Candace Condry for helping to manage the study at the
animal shelter. Maryl`ene Gagn´e was supported by a doctoral fellowship from the Social Sciences and
Humanities Research Council of Canada and by an American Psychological Association Dissertation
Award.
2Department of Management, John Molson School of Business, Concordia University, Montreal,
Canada.
3Address all correspondence to Maryl`ene Gagn´e, Department of Management, John Molson School
of Business, Concordia University, GM 503-49, 1455 de Maisonneuve blvd West, Montreal, Quebec,
Canada H3G 1M8; e-mail: mgagne@jmsb.concordia.ca.
199
0146-7239/03/0900-0199/0 C
°2003 Plenum Publishing Corporation
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
200 Gagn´e
If people are good only because they fear punishment, and hope for reward, then
we are a sorry lot indeed Einstein
There is no doubt, as Einstein implied, that good deeds are essential to the pro-
motion of a healthy society. Perhaps for this reason, social psychologists have
been interested in the questions of why and under what conditions people will
act to benefit other people, actions that are called prosocial (Staub, 1978). Much
social psychological research has examined the conditions that facilitate or inhibit
helping behaviors, such as bystander intervention (Latan´e & Darley, 1970), volun-
teering(Batson et al., 1997), and giving(Cialdini & Kenrick, 1976). Some research
also addresses people’s motives for engaging in certain prosocial behaviors, such
as volunteering (e.g., Clary et al., 1998). Some interesting findings have advanced
our knowledge about the effects of these factors, but still much research is needed
to uncover what motivates prosocial behavior.
In an attempt to increase likelihood of engagement in prosocial behavior,
various writers have called for the use of incentives (Skinner, 1978), and for mak-
ing prosocial behavior “mandatory” within certain settings (Sobus, 1995). Incen-
tive systems would seem at first glance to provide an easy solution for increasing
prosocialbehavior.Indeed,suchsystemsarealready in wide use with,forexample,
benefactorplates,donordinners,andgifts to peoplewhocontributetocharities.Un-
fortunately, research on the effects of tangible incentives on prosocial behavior has
foundthatincentivesmaynot be themosteffectivemeansofmaintaining or increas-
ing prosocial behavior and, under certain conditions, can even deter such behavior.
For example, Fabes, Fultz, Eisenberg, May-Plumlee, and Christopher (1989)
found that rewarding children for helping sort pieces of paper for sick children
undermined their subsequent helping behavior. This study showed that, although
incentives might increase helping while they are in effect, they can decrease subse-
quenthelping,perhaps because the enactment ofhelping behavior becomes contin-
gent on these recently introduced incentives. A few studies showed that incentives
can, in fact, alter people’s perceptions of themselves as acting altruistically. Kunda
and Schwartz (1983) found that payments undermined undergraduates’ internal-
ized sense of moral obligation to help a blind person upon request, and this in
turn deterred helping behavior toward this person. Batson, Coke, Jasnoski, and
Hanson (1978) similarly found that people who were paid to help an experimenter
code data judged themselves to be less altruistic than people who were not paid
for helping or who did not help. Parallel findings have been found in applied set-
tings. Upton (1974) demonstrated that committed blood donors who received a
reward were subsequently less likely to donate compared to committed donors
who were not offered a reward. Finally, a study by Paulhus, Shaffer, and Downing
(1977) showed that people’s stated willingness to donate blood in the future was
lower when blood-drive workers stressed selfbenefits from donating, as opposed
to stressing the good will of donors.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 201
Even some economists, whose theories mostly focus on human beings as
being solely motivated by economic gains, are acknowledging that strict con-
tingency systems can sometimes deter some forms of behavior, such as volun-
teering and giving blood. For instance, Frey (1993) reviewed research on the
effectiveness of incentive and sanction systems on behaviors like environmental
conservation and blood giving, concluding that such systems are ineffective for
activities that most people would do for the value they find in them. He reported
that people have generally negative attitudes about their blood being purchased
and that the introduction of payments to regular donors decreased their donation
rates.In some cases, people have attempted to increase prosocial behavior through
the use of other external contingencies without much success. In a compelling
article on “mandatory volunteering” (which is an oxymoron), Sobus (1995) ex-
plored the effectiveness of high school programs that have for mission to de-
velop adolescents’ awareness of community needs and altruistic inclinations.
Inherent in these programs is a lack of choice about participating, salient eval-
uations to meet school requirements, and deadlines for work completion. As
will be discussed later, all of these factors have been shown to decrease enjoy-
ment and engagement in certain activities (Amabile, DeJonc, & Lepper, 1976;
Inagaki & Hatano, 1984; Zuckerman, Porac, Lathin, Smith, & Deci, 1978). In-
terviews with students revealed generally negative attitudes toward mandatory
volunteering programs and found that forcing volunteer work defeated the pur-
pose of fostering moral or prosocial behavior. Students felt that they were not
giving anything through their behavior, because they were forced into it. More-
over, the general public did not hold a positive view of students’ volunteering
behavior, because they did not see it as altruistically motivated but instead as
forced through social pressure. Sobus consequently argued that self-determination
is essential for students to internalize the value of their citizenship behavior,
and that forcing them to engage in such behavior robs them of this sense of
self-determination. A recent study supports this argument by showing that stu-
dents required to volunteer evidenced a decrease in future intentions to volun-
teer compared to students who were given a choice (Stukas, Snyder, & Clary,
1999).
Taken together, these studies indicate that prosocial behavior, be it helping
others through volunteering or through giving blood, seems to be negatively af-
fected when people feel obligated or controlled by external contingencies. The
issue of acting prosocially either volitionally or through external forces can be
examined with a theoretical framework that addresses how environmental forces
and individual differences can affect motivation to engage in these behaviors. In
the present analysis, I tested a model derived from self-determination theory (Deci
& Ryan, 1991) that postulates that the satisfaction of a person’s core psychologi-
cal needs is a requirement to the motivation, especially long-term motivation, of
activities that are prosocial.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
202 Gagn´e
SELF-DETERMINATION THEORY
Self-determination theory (Deci & Ryan, 1985a, 1991, 2000) proposes that
human beings have basic psychological needs for autonomy, competence, and
relatedness. Contexts that support the satisfaction of these needs will promote a
person’s enjoyment of activities and the autonomous self-regulation of behaviors.
People are more likely to be intrinsically motivated, that is, to do an activity simply
for the enjoyment they derive from it, when they can freely choose to pursue the
activity (autonomy), when they master the activity (competence), and when they
feel connected and supported by important people, such as a manager, a parent, a
teacher, or team-mates (relatedness). Early research using this framework focused
mostly on examining how decreases in experienced autonomy influenced intrin-
sic motivation. For example, research has shown that controlling rewards (Deci,
1971; Deci, Koestner, & Ryan, 1999), deadlines (Amabile et al., 1976), and eval-
uation (Amabile, 1979) can decrease the enjoyment of an activity, whereas choice
(Zuckerman et al., 1978) and acknowledging people’s feelings toward activities or
rulesregardingan activity (Koestner,Ryan, Bernieri,& Holt, 1984)can enhance it.
The assumption in the present research is that motivation for prosocial be-
havior can also be dampened by these controlling factors, and enriched by fac-
tors like choice and acceptance, because these factors affect the satisfaction
of basic psychological needs. Although motivation is not directly measured in
the present studies, I measured the satisfaction of the three psychological needs,
which is theorized to promote volitional motivation (Deci & Ryan, 2000).
Ryan and Deci (2000) argued that we are naturally inclined to be prosocial
animals, given proper nurturing. When we lack this nurturing, we are likely to
substitute it by pursuing goals that might appear on the surface to satisfy basic
psychological needs (e.g., acquiring status or financial success; Kasser & Ryan,
1993) but that do not promote prosocial behavior. This means that when our basic
psychological needs are unfulfilled, we are more likely to engage in behaviors
that have ourselves as the focus. One longitudinal study in particular found that
adolescents with cold and controlling mothers were likely to value financial suc-
cess over community values, compared to adolescents with warm and supportive
mothers,evenwhen controlling for themothers’ own values (Kasser, Ryan, Zax, &
Sameroff, 1995). Kasser and Sheldon (2000) also found that participants induced
to feel a general sense of insecurity acted more greedily and consumed more
resources in a forest-management game. Thus, the rationale behind the present
studies is that need satisfaction should orient people toward paying more attention
to others, thus making them more likely to engage in prosocial behavior.
Another goal of the studies is to examine factors that would affect need sat-
isfaction. In this regard, self-determination theory proposes that both individual
differences in autonomy orientations and contextual supportiveness will influence
need satisfaction. First, I will address the issue of individual differences in auton-
omy orientations.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 203
Individual Differences in Autonomy Orientation
People’s early social experiences can influence the development of their
causality orientation, thus making people vary in terms of how they are ori-
ented toward being autonomously self-regulated or being regulated by the en-
vironment (i.e., controlled). Autonomously regulated people feel agentic in their
own behavior, whereas controllingly regulated people feel like pawns to external
forces(deCharms, 1968). The GeneralCausality Orientations Scale(Deci & Ryan,
1985b) assesses three different causality orientations to action: (1) an autonomous
orientation representing a tendency towards volitional engagement in action (in-
ternal locus of causality); (2) a controlled orientation representing a tendency to
orient toward and to be regulated by controls and contingencies (external locus of
causality); and (3) an impersonal orientation representing a tendency not to engage
in action, akin to helplessness (impersonal locus of causality). Independent of how
supportive the context is, differences in causality orientations can lead people to
have their basic needs differentially met, because people with different causality
orientations may perceive the same context differently, and/or because people with
different orientations may elicit different reactions during interpersonal interac-
tions (Deci & Ryan, 1985a).
Autonomyorientation has beenrelatedtovariouspositiveoutcomes. Williams,
Grow, Freedman, Ryan, and Deci (1996) found that morbidly obese patients’
causality orientations predicted the adoption of more autonomous motivation for
following a diet, which in turn was related to the amount of weight they lost.
Williams, Freedman, and Deci (1998) found similar results for diabetic patients’
glucose control. Finally, Baard, Deci, and Ryan (in press) found that employees
with a high autonomy orientation experienced higher need satisfaction at work,
which in turn increased their performance evaluations and well-being. Because
autonomy orientation has been shown to influence need satisfaction, it is likely
to influence prosocial behavior engagement. Moreover, research on parenting has
shown that the amount of nurturance a person receives early in life influences the
developmentof empathic responding, andempathy has been related to engagement
in prosocial behavior (Eisenberg-Berg, 1979; Hoffman, 1984; Koestner, Franz, &
Weinberger, 1990).
The Support of Autonomy
As stated earlier, contextual factors can affect need satisfaction as well.
Contexts that are described as autonomy supportive are characterized as giving
people choice and encouragement for personal initiative and also support peo-
ple’s competence in a climate of relatedness (Deci et al., 2001) are predicted to
promote autonomous motivation (e.g., intrinsic motivation) as opposed to con-
trolled motivation (e.g., extrinsic motivation). Concretely speaking, the quality of
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
204 Gagn´e
interaction with significant others, such as teachers, managers, or parents, can af-
fect the degree to which an individual feels autonomous, competent, and related,
and this can affect the degree to which he or she comes to value and even enjoy
an initially uninteresting goal or activity. An autonomy supportive person (or a
work environment) would typically provide a good rationale for asking someone
to engage in an activity, give some choice to the person, acknowledge the person’s
feelings toward the activity, and encourage the person to take initiative and convey
confidence in the person’s abilities (Williams, Gagn´e, Ryan, & Deci, 2002).
A substantial body of research has confirmed the importance of autonomy
support in promoting positive outcomes. For instance, autonomy support has been
shown to lead to greater engagement in an initially uninteresting activity and
increased positive feelings toward the activity (Deci, Eghrari, Patrick, & Leone,
1994). This effect was replicated in elementary schools, where teacher and parent
autonomysupport were shown to relateto better academicperformance and school
adjustment in children (Grolnick & Ryan, 1989; Ryan, Stiller, & Lynch, 1994).
WilliamsandDeci(1996)found that supervisory autonomy supporthelpedmedical
students’ development of psychosocial values and fostered autonomy supportive
behavior toward patients. In the work domain, Deci, Connell, and Ryan (1989)
trainedmanagers to be autonomysupportivewith their subordinates andfound that
autonomy support predicted later trust in the organization, positive affect at work,
and work satisfaction. Gagn´e, Koestner, and Zuckerman (2000) also found that
management autonomy support predicted increased acceptance of organizational
change over a 13-month period.
Somestudieshaveshownthat the link between autonomy support and positive
outcomes is mediated by need satisfaction. Grolnick, Ryan, and Deci (1991) found
thatparentalautonomysupport wasrelated to children’sperceived competence and
autonomy at school, which in turn was related to academic performance. Baard
et al. (1999) and Deci et al. (2001) found support for a model where management
autonomy support was related to the satisfaction of employees’ needs, which was
related in turn to higher performance evaluations, engagement in one’s work, and
well-being, in both Bulgarian and American samples. The results obtained in or-
ganizational studies might have some implications for the domain of volunteering,
where work climate is likely to be an important source of need satisfaction.
Autonomysupport may also help preventnegativeoutcomes,suchasturnover.
Vallerand, Fortier, and Guay (1997) found that students’ perceptions of teacher,
parent, and school administration autonomy support was related to greater feelings
of competence and autonomy, which in turn promoted autonomous motivation.
Autonomousmotivationin turn was negativelyrelated to intentions to drop out, and
to actual drop out behavior. Parallel results were found in organizational studies
on turnover. For example, Rhoades, Eisenberger, and Armeli (2001) found that
perceived organizational support was negatively related to voluntary turnover, and
Frone (2000) found that conflict with supervisors led to higher turnover intentions.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 205
Overview of the Studies and Hypotheses
Iconductedtwostudiesto test a model derivedfromself-determination theory
whereby autonomy support and autonomy orientation are directly related to need
satisfaction, which in turn is directly related to engagement in prosocial activities.
In Study 1, undergraduate students completed questionnaires on their habitual en-
gagement in 10 different prosocial activities, and completed measures of parental
autonomy support and need satisfaction. It was hypothesized that students with
a high autonomy orientation and who perceive their parents as being autonomy
supportive would report higher need satisfaction in their life, and this would be
related to extent of their engagement in the prosocial activities. In Study 2, volun-
teer workers at an animal shelter completed a questionnaire on their perceptions of
work autonomy support, need satisfaction, and psychological engagement in their
work, and I obtained the number of hours subsequently worked for the subsequent
4 months, and whether volunteers quit or not. It was hypothesized that volunteers
with a high autonomy orientation and who perceive the work environment to be
autonomy supportive would report higher work need satisfaction, and this would
predict the extent to which they are psychologically engaged in their volunteer
work, and the number of hours they volunteer during the 4-month period. It was
also hypothesized that an autonomy supportive work environment would reduce
the likelihood of volunteer turnover, with the effect mediated by need satisfaction.
STUDY 1
The first study tested the self-determination model by asking college students
to report on their prosocial activities, and to answer questionnaires pertaining to
theirlevelof autonomy orientation, their parents’autonomy support, and their need
satisfaction. Autonomy orientation and autonomy support were hypothesized to
have a positive influence on engagement in prosocial activities. General need satis-
faction was predicted to mediate the relations of autonomy support and autonomy
orientation on engagement.
Method
Participants and Procedure
One hundred twenty-one (77 women, 42 men, 2 unreported) undergraduate
psychology students participated in a questionnaire study titled “preferences for
social activities” for extra credit. One participant’s data were not used because the
questionnaire was not properly completed. Students completed the questionnaire
in group sessions, following informed consent, and were fully debriefed after
completion.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
206 Gagn´e
Measures
Themeasureswere included in the following order.TableI presents the means
and standard deviations for each variable.
Reports on Prosocial Activities. Respondents reported if they have ever, or
currently engage, in the following behaviors, on a scale from 1 (never)to7(very
often): volunteer for a nonprofit organization, donate money to a charitable or-
ganization, vote, sign petitions, recycle, give in food drives, help in emergency
situations, actively support causes (activism), donate blood, and give away furni-
ture and clothes. Although these are 10 very different forms of behavior requiring
different levels of involvement and personal costs, and yielding different social
rewards, it is assumed that the same psychological processes would be involved
in motivating them. A reliability analysis showed that the reports of engagement
in these different activities were internally consistent, α=.69. Therefore, reports
were averaged to form a single index of engagement in prosocial behavior.
General Causality Orientations Scale. This scale (Deci & Ryan, 1985b)
measured students’ general orientations toward being autonomous (Cronbach
α=.82), controlled (α=.73), and amotivated (impersonal causation; α=.85).
Respondents read 12 hypothetical scenarios and rated the likelihood of having
each of three possible reactions depicting autonomous, controlled, and impersonal
orientations to each scenario ona1(very unlikely)to7(very likely) scale. The
control and impersonal subscales were not used in this study.
General Need Satisfaction Scale. This scale was adapted from a measure of
need satisfaction at work (Ilardi, Leone, Kasser, & Ryan, 1993). Respondents
indicated on a scale from 1 (not true at all)to7(definitely true) the extent
to which the psychological needs of autonomy (7 items, α=.69), relatedness
(6 items, α=.86), and competence (8 items, α=.71) are generally satisfied in
their life. Examples of items are, “I feel like I can decide for myself how to live
my life” (autonomy), “I really like the people I interact with” (relatedness), and “I
often do not feel very capable” (competence, reversed). To test the model, the three
subscales were averaged to form an index of general need satisfaction (α=.89),
as their correlations were between .61 and .66. This method has been used in many
other studies (e.g., Deci et al., 2001), but correlations of each subscale with other
variables are still reported in Table I.
Parental Autonomy Support Scale. Respondents rated this 9-item scale
(Robbins, 1994) measuring mother’s (α=.90) and father’s (α=.89) autonomy
support ona1(not true at all)to7(definitely true) scale. Examples of items are,
“Mymothertries to tell me how to run mylife” (reversed),and“My father helps me
choose my own direction.” Scores for mother and father (r[112] =.35, p<.001,
α=.89) were averaged to form a parental autonomy-support index. Since the cor-
relation between the subscales was not very high, correlations of other variables
with the two subscales are presented separately in Table I.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 207
Table I. Means, Standard Deviations, and Correlations for Variables in Study 1 (N=118)
Mean (SD)123 4 5678
1. Autonomy orientation 5.63 (0.75)
2. Mother autonomy support 5.44 (1.15) .17
3. Father autonomy support 5.07 (1.30) .22∗.48∗∗∗
4. Parental autonomy support 5.25 (1.05) .23∗∗ .84∗∗∗ .88∗∗∗
5. Autonomy need 5.00 (0.80) .27∗∗ .27∗∗ .51∗∗∗ .46∗∗∗
6. Competence need 4.97 (0.93) .35∗∗∗ .32∗∗∗ .46∗∗∗ .45∗∗∗ .66∗∗∗
7. Relatedness need 5.60 (1.20) .51∗∗∗ .35∗∗∗ .54∗∗∗ .53∗∗∗ .61∗∗∗ .64∗∗∗
8. Need satisfaction 5.19 (0.80) .45∗∗∗ .36∗∗∗ .58∗∗∗ .56∗∗∗ .85∗∗∗ .88∗∗∗ .88∗∗∗
9. Prosocial engagement 3.30 (0.94) .35∗∗∗ .05 .22∗.16†.19∗.40∗∗∗ .26∗∗ .34∗∗∗
†p<.10. ∗p<.05. ∗∗ p<.01. ∗∗∗ p<.001.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
208 Gagn´e
Results and Discussion
Complete data were obtained from 118 students who reported engaging in
a median of 7 out of 10 activities. Mean engagement ranged from 2.03 to 5.46
across the 10 activities, with an overall average of 3.3. Correlations between the
10 individual activities and the other variables were rather low (between. 06 and
.20) but considering individual activities ratings as the unit of analysis is like
using single items. Correlations between the variables are presented in Table I,
where it is shown that the composite of engagement in prosocial activities is more
strongly correlated with the other variables than were the individual activities.
Autonomy orientation and autonomy support were significantly, but not highly
correlated. Autonomy orientation was strongly related to prosocial engagement.
Interestingly, father autonomy support was more strongly correlated with engage-
ment than mother autonomy support. This parallels Koestner et al.’s finding that
paternal involvement in child care was a predictor of empathy development in a
26-year longitudinal study (Koestner et al., 1990). However, to keep in line with
previous research (e.g., Vallerand et al., 1997), the combined measure of parental
autonomysupport was used, which meansthat the reported resultsare conservative
estimates of the effect of autonomy support. As expected, each need satisfaction
subscale was strongly correlated with autonomy orientation and parental auton-
omy support. They were also substantially correlated with prosocial engagement,
with a stronger correlation for competence. Nonetheless, to be consistent with
previous research (e.g., Deci et al., 2001), I used the composite of all three needs.
Overall, correlations provide good initial support for the mediational model to be
tested through path analysis, except for the low correlation of autonomy support
with prosocial engagement.
Testing the Hypothesized Mediational Model
To test the postulated model, including a test of the mediating effects of
need satisfaction, I conducted a series of path analyses with Amos 4 (Arbuckle,
1999) with maximum-likelihood estimation. Because the complete sample avail-
able to run the analysis was small (N=118), scores on subscale items were
averaged to create observed variables (represented by rectangles in the figures),
and the variance of the error terms was fixed with the subscale reliabilities
(Cronbach α’s) and standard deviations (Bollen, 1989; Wang, Fisher, Siegal,
Falck, & Carlson, 1995), thus creating the equivalent of latent variables (rep-
resented by ellipses in the figures). This reduced the maximum number of es-
timated parameters to 8, creating an adequate ratio of cases to parameter of
14.75 to 1.
To test for mediation, indirect effects were calculated for each predictor vari-
able. The total effect of a predictor variable on an outcome variable is the sum of
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 209
the direct effect of the predictor on the outcome and its indirect effect through a
mediator variable, so a significant indirect effect indicates that the mediator sub-
stantially reduced the direct effect of the predictor on the outcome variable. To
calculate the indirect effect of the predictor on the outcome, one needs to subtract
the direct effect from the total effect (Allison, 1999). Also, Baron and Kenny’s
method (Baron & Kenny 1986), adapted for structural equations modeling, was
used (Hoyle & Smith, 1994).
The first model tested the direct paths from autonomy orientation and from
parental autonomy support to prosocial engagement. Because the correlation be-
tween autonomy orientation and autonomy support was substantial (see Table I),
which is not surprising considering they are measured at the same level (Vallerand,
1997), I freed the covariance between the two variables. However, this model
was just identified, meaning there were no degrees of freedom left to assess
its fit to the data. Fixing this covariance to 0 yielded a poor fit, as can be seen
from these inconsistent fit indices, χ2(1) =6.49, p<.01, GFI =0.97, NNFI =
0.15, CFI =0.72, RMSEA =0.22. The unstandardized path coefficient from au-
tonomy orientation to prosocial engagement was significant, γ=0.51, p<.01,
but the one from parental autonomy support to engagement was not, γ=0.08,
ns. The second model added need satisfaction as a partial mediator for both
autonomy orientation and parental autonomy support leading to prosocial en-
gagement (keeping the direct paths). Again, this model was just identified, so it
was not possible to assess its fit. Fixing the covariance between autonomy ori-
entation and autonomy support to 0 yielded a poor fit, χ2(1) =6.49, p<.01,
GFI =0.97, NNFI =0.61, CFI =0.93, RMSEA =0.22, and the direct path
from autonomy support to engagement was close to 0, while the others were all
significant.
Afterexaminingmodification indices, the direct pathfrom parental autonomy
support to prosocial engagement was dropped, thus modeling full mediation of
need satisfaction for this path. The unstandardized results of this final model are
presented in Fig. 1. The goodness of fit of this model was excellent, χ2(1) =
0.28, ns, GFI =1.00, NNFI =1.05, CFI =1.00, RMSEA =0.00. All estimated
pathswere significant. Both parentalautonomy support andhaving an autonomous
orientation influenced general need satisfaction, and need satisfaction positively
influenced engagement in prosocial activities. Autonomous orientation retained
a direct path to engagement, but was also mediated through need satisfaction, as
indicated in a significant indirect effect of .11, p<.05. The effect of parental
autonomy support on engagement was fully mediated by need satisfaction, with
an indirect effect of .10, p<.05. Thus, it appears that parental autonomy support
had a small effect (partial R2=.05) on engagement, and this effect was fully
mediated by need satisfaction. Having an autonomous orientation had substantial
influences on engagement in prosocial activities, partly through increasing need
satisfaction.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
210 Gagn´e
Fig. 1. Results of the final path analytic model tested in Study 1.
STUDY 2
The design of the first study was limited in at least two ways. First, prosocial
behaviors were self-reported and retrospective. To remedy this, it would be neces-
sary to observe actual prosocial behavior in order to provide some external validity
for the model. Second, parental autonomy support is a relatively distal antecedent
of prosocial motivation for adults, including undergraduate students. Although
parental autonomy support is likely to have long-term and wide-ranging effects in
many aspects of people’s lives, more proximal sources of autonomy support might
have greater effects on domain-specific behavior. For example, supervisor auton-
omy support might have potent effects on the motivation of workers. Moreover,
parental autonomy support’s influence on prosocial behavior might be confounded
with the fact that autonomy-supportive parents could be the kind of people who
engage in prosocial activities themselves. The effects of modeling such behavior
for their children could influence their children’s behavior, just as much as the
effect of parents being generally supportive of their children’s behavioral choices.
The second study addressed these two limitations. Volunteers working in an
animal shelter were asked to complete a questionnaire. The volunteer coordinator
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 211
provided the number of hours worked by each volunteer during the 4 months
that followed the administration of the questionnaire. Thus, the first limitation of
Study 1 was addressed by observing actual prosocial behavior, that is, the number
of hours spent at the shelter. Moreover, since assessment of prosocial behavior
was done after volunteers completed the questionnaire, it was possible to look
at the prospective correlations between perceived autonomy support and engage-
ment in volunteer behavior. This is considered to be a measure of the quantity of
engagement. A measure of the quality of the engagement was added with a scale
measuring psychological engagement in work activities. Finally, I asked volun-
teers, after the collection of data, to report whether they were still volunteering at
theshelter4monthslater.These data permitted to test the hypothesis that autonomy
support might help curtail turnover.
The questionnaire contained scales assessing the autonomy supportiveness of
the climate in this volunteer work setting, along with scales assessing individual
differences in motivational orientations, need satisfaction at the workplace, and
feelings of engagement in volunteer work. Thus, the second limitation of Study 1
was addressed by assessing the perceived autonomy supportiveness of the context
in which the volunteer behavior occurs. These improvements permitted a more
rigorous test of the model.
Method
Participants and Procedure
Two hundred twenty-seven volunteers, 42 men and 185 women, from a Hu-
mane Society for Animal Welfare situated in the Northeastern U.S. were recruited
via mail to participate in a questionnaire study about their experience as volun-
teers. This animal shelter is a state-run nonprofit organization that serves to control
domestic animal populations, enforce laws concerning animal abuse, and educate
the community about animal welfare. Some workers at the shelter were paid em-
ployees and others were volunteers, but only the volunteers participated in this
research. Their tasks varied greatly, ranging from cleaning cages and pens, to so-
cializing animals, to greeting potential adopters, to participating in pet therapy
programs.
An envelope containing a letter explaining the purpose and procedure of the
study, a consent form, and the initial questionnaire, was mailed to each volunteer.
After receiving the volunteers’ consent and their completed questionnaires, the
number of hours each participant had worked during the 4 months following the
administration of the questionnaire was obtained from the volunteer coordinator.
At the end of this period, volunteers were sent a short questionnaire to know
if they were still volunteering and other questions not of interest to the present
study. In order to ensure the confidentiality of the responses, volunteers were
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
212 Gagn´e
given identification numbers to match their questionnaire to the list containing the
number of hours that each volunteer worked. A technical report of the aggregated
results was presented to the shelter.
Measures
The questionnaire started with questions concerning demographic character-
istics, including age, gender, education level, income, and whether the respondents
engagedinothervolunteeractivities in their lives.Whensentthesecondshortques-
tionnaire, volunteers were asked to report on whether they were still volunteering,
and also to approximate the number of hours they had worked in the last 4 months.
If, for clerical reasons, number of hours could not be obtained for particular vol-
unteers (e.g., some volunteers participated in a pet therapy program for which
the coordinator did not record hours), self-reported hours spent volunteering was
used as a behavioral measure of prosocial behavior. The correlation between self-
reported and the number of hours reported by the shelter for the 58 volunteers
for whom both measures were available was .80. Respondents were also asked to
complete a series of measures in the following order, some of which were modi-
fied to pertain to volunteer work. All of the questions were phrased in terms of the
volunteers’ experiences at the shelter during the preceding 6 months.
GeneralCausalityOrientations Scale. This is the samescale used inStudy 1.
Again, only the autonomy orientation subscale was used, α=.82.
Work Climate Survey. Items from the first two sections of this survey (Deci
et al., 1989), which was modeled after the Job Diagnostic Survey (Hackman &
Oldham,1975),were used to assess the volunteers’perceptionsof the work climate
at the shelter. Thirteen items, consisting of five subscales, measured perceptions of
autonomy support. The first three subscales consisted of statements that workers
rated ona1(very inaccurate)to7(very accurate) scale. Three items concerned the
immediate supervisors (e.g., The staff at the shelter give me a great deal of choice
about how to do my volunteer work and how to handle problems I encounter;
α=.73), three items concerned the shelter management in general (e.g., The
managersat the shelter consider carefully the impact of decisions on the volunteers’
lives; α=.84), and three items concerned the structure of the work environment
(e.g., At the shelter, there are guidelines and regulations that let me know what
to do in almost every situation; α=.68). The environmental structure subscale is
not always included in the autonomy support composite, but because of the nature
of operations at the shelter, which require providing some structure to volunteers
who are typically there only a few hours a week and need some direction to
accomplish their work, it was included as an indicator of autonomy support in
the present study. The other two subscales consisted of adjectives describing the
work atmosphere rated ona1(not at all)to7(extremely) scale. The first one
contained four adjectives describing the supportiveness of the work environment
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 213
(e.g., supportive; α=.89), and the second contained seven adjectives describing
how people feel in the environment (e.g., flexible; α=.91). Items within each of
the subscales were averaged, and then the five subscales were averaged to create
a variable labeled work autonomy support, α=.85. Correlations between the
subscales ranged from .23 to .74.
Need Satisfaction Scale. Twenty items assessed participants’ experiences of
psychological need satisfaction at work (Ilardi et al., 1993). They indicated on a
scale from 1 (not true at all)to7(definitely true) the extent to which they felt
satisfaction of the needs for autonomy (7 items; α=.76), relatedness (6 items;
α=.81), and competence (8 items; α=.60). Examples of items are, “I feel like I
canmake a lot ofinput todeciding how myvolunteer work gets done”(autonomy),
“I really like the people I volunteer with” (relatedness), and “I do not feel very
competent when I am volunteering” (competence, reversed). Items within each
subscale were averaged, and the three subscales were then averaged to form a
variable labeled need satisfaction at work (α=.88). Correlations between the
subscales ranged from .62 to .65.
Work Engagement Scale. This 12-item scale (Deci et al., 2001; α=.74)
measured self-reported behavioral and emotional engagement in volunteer work
ona1(completely disagree)to7(completely agree) scale. Sample items are,
“When I am volunteering, I work as hard as I can” (behavior) and “When I am
volunteering, I often feel bored” (emotion, reversed). The 12 items were averaged
to form a measure of psychological engagement.
Results and Discussion
One hundred volunteers, 22 men and 78 women, completed and returned the
firstquestionnaire, which yielded aresponse rateof 44%. Anotherrecent volunteer
study obtained a response rate of 32% (Penner & Finkelstein, 1998), so the present
response rate seems typical. The data from one volunteer were discarded because
the questionnaire was not properly completed. Of the 99 volunteers who were sent
the short questionnaire at the end of the 4-month period, 60 returned it, 2 reported
continuingto volunteerbutdid not completethequestionnaire,17reported that they
had stopped volunteering so did not complete the questionnaire, and 20 provided
neither the questionnaire nor other information. Volunteers who completed both
questionnaires or at least reported that they stopped volunteering were older and
more educated than volunteers who completed only the first questionnaire. None
of the demographic variables interacted with any of the variables in the model and
they will therefore not be discussed further.
The average amount of time spent volunteering during the 4-month period
was obtained by the coordinator for 144 volunteers, and the mean amount of
volunteered time was 29.49 hr, or an average of 7.37 hr per month. Age, education
level, income level, and tenure were unrelated to the amount of time volunteered.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
214 Gagn´e
A 2 (participated vs. did not) ×2 (gender) between-subjects analysis of variance
was conducted in order to evaluate if there were differences in the amount of time
volunteered. There was only a main effect for participation, such that volunteers
who participated in at least the first assessment spent more time volunteering
(M=34.77 hr) than volunteers who did not participate in either assessment (M=
22.31 hr), F(1,140) =4.12, p<.05, r=.17. There wasno significant difference
inthe number of hoursbetween those whoself-reported and thosewhose data were
reported by the volunteer coordinator.
Table II presents the means and standard deviations of the study variables,
as well as correlations between the variables. First, it is interesting to note that
psychological engagement and the number of worked hours were not significantly
correlated (although in the predicted positive direction), which may mean that
quality and quantity of engagement are distinct constructs and could be differen-
tially affected by predictor variables. Autonomy support was marginally related to
thenumber of volunteered hoursand was not relatedto psychologicalengagement.
It was also positively related to need satisfaction. Autonomy orientation was not
related to the number of volunteered hours, but was significantly related to psycho-
logical engagement. In other words, autonomy orientation influenced the quality
of engagement, but not quantity. Autonomy orientation was also not strongly cor-
related with need satisfaction, a finding that was significant in Study 1. This could
be explained by the fact that need satisfaction was assessed at a different level of
analysis: In Study 1, a measure of general life need satisfaction was used, whereas
in Study 2, it was a measure of work-specific need satisfaction. Nevertheless, need
satisfaction was related to both quality and quantity of engagement. Looking at the
separate correlations for each subscale, correlations with the competence subscale
were stronger than for the other subscales, replicating findings in Study 1.
Testing the Hypothesized Mediational Model
A series of models similar to the ones in Study 1 was tested using path
analysis with Amos 4 (Arbuckle, 1999) with maximum-likelihood estimation.
Eighty-one cases had complete data for this analysis. As in Study 1, scores on
subscale items were averaged to create observed variables, and variance of the
error terms was fixed using subscale reliabilities and standard deviations, except
for the number of hours, which was an observed variable without measurement
error. This reduced the maximum number of estimated parameters to 10, creat-
ing an adequate ratio of cases to parameter of 8.1 to 1. Again, I started with a
model with direct paths from autonomy orientation and autonomy support to psy-
chological engagement and hours worked. This time, the covariance between the
predictors was set to 0, as they were not correlated (see Table II), which is to be
expected given that they are measured at different levels of analysis (Vallerand,
1997). This model yielded a modestly adequate fit, χ2(2) =3.23, ns, GFI =0.98,
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 215
Table II. Means, Standard Deviations, and Correlations for Variables in Study 2 (N=81)
Mean (SD)1 234567
1. Autonomy orientation 5.85 (0.67)
2. Work autonomy support 4.96 (1.20) .04
3. Autonomy need 5.16 (1.05) .08 .73∗∗∗
4. Competence need 5.44 (1.02) .15 .60∗∗∗ .65∗∗∗
5. Relatedness need 5.23 (1.13) .11 .73∗∗∗ .63∗∗∗ .61∗∗∗
6. Need satisfaction 5.28 (0.93) .13 .79∗∗∗ .87∗∗∗ .86∗∗∗ .87∗∗∗
7. Psychological engagement 5.94 (0.74) .35∗∗∗ .16 .30∗∗ .50∗∗∗ .28∗.41∗∗∗
8. Number of hours 35.26(35.32) −.10 .20†.17 .31∗∗ .27∗.29∗∗ .17
†p<.10. ∗p<.05. ∗∗ p<.01. ∗∗∗ p<.001.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
216 Gagn´e
Fig. 2. Results of the final path analytic model tested in Study 2.
NNFI =0.73, CFI =0.91, RMSEA =0.09. The path from autonomy orientation
to psychological engagement was significant, γ=0.46, p<.01, and the one to
hours worked was not, γ=0.11, ns. The path from autonomy support to hours
workedwas marginallysignificant,γ=7.31, p=.11,whiletheoneto psycholog-
ical engagement was not, γ=−6.02, ns. Next, a model adding need satisfaction
as a partial mediator, keeping the two direct paths that were significant, was tested.
This model yielded a comparable fit to the first model, χ2(4) =10.14, ns, GFI =
0.94,NNFI =0.80, CFI =0.92, RMSEA =0.17.All paths were significant,except
for the one from autonomy support to hours worked. So a final model was tested
removing this path. Unstandardized results are presented in Fig. 2. The model
achieved an acceptable fit, χ2(5) =10.30, ns, GFI =0.94, NNFI =0.85, CFI =
0.93, RMSEA =0.14.
In the final model, work autonomy support was positively related to need
satisfaction at work, but autonomy orientation was not. Need satisfaction in turn
predicted both psychological engagement and the number of volunteered hours.
Work autonomy support had an effect on psychological engagement that was me-
diated by need satisfaction, as indicated by an indirect effect of .20, p<.05.
However, since the correlation between autonomy support and engagement was
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 217
not significant, it is difficult to interpret this result. Perhaps this effect is caused
by the high correlation between autonomy support and need satisfaction, in con-
junction with the high correlation between need satisfaction and psychological
engagement. Work autonomy support also positively influenced the number of
volunteered hours through its effect on need satisfaction (indirect effect =8.27,
p<.05). Since there was a marginally significant correlation between autonomy
support and worked hours, there is stronger support for a mediating effect. Auton-
omy orientation had a significant direct influence on psychological engagement,
but did not have a significant indirect effect through need satisfaction, .05, ns.
Autonomy orientation did not have any effect (direct or indirect) on worked hours.
Thus, the hypothesized model was partially supported in that autonomy orienta-
tion predicted engagement, but not through need satisfaction, whereas autonomy
support predicted engagement through need satisfaction.
Predicting Volunteer Turnover
To test the hypothesis that autonomy support would influence turnover, an
independent-samples t-test was conducted in order to examine whether there were
differences in work autonomy support between volunteers who reported quitting
(n=17) versus those who reported continuing to volunteer (n=62). There was
a significant difference in perceived work autonomy support, t(75) =2.04, p<
.05, r=.23,4such that those who quit perceived the work climate to be less
autonomy supportive (M=4.39) than those who did not (M=5.09). The same
analysis was conducted using the other variables in the model and demographic
information as dependent variables, but all yielded nonsignificant results. These
resultssuggestthatvolunteerswhoquit during the course of the studyperceivedthe
work environment to be less autonomy supportive than volunteers who continued
to volunteer at the shelter. No other differences were found between quitters and
nonquitters.
To examine if autonomy support predicted whether someone quit volunteer-
ing or not, a binary logistic regression analysis with quitting as the dependent
variable was conducted, entering work autonomy support as a predictor. Work au-
tonomy support significantly improved correct classification (76.62%) of quitters
and nonquitters over no predictors, χ2(1) =3.93, p<05, and its nonstandardized
beta coefficient was −.43, pr =−.14, indicating that lower perceived autonomy
support predicted higher likelihood of quitting. Need satisfaction was entered next
in the equation, but yielded no significant effect. Thus, the hypothesis was partially
supported in that autonomy support was shown to predict turnover, but the effect
was not mediated by need satisfaction.
4A measure of effect size, the Pearson r, was computed as, r=qF
F+dfe(Rosenthal & Rosnow,
1984).
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
218 Gagn´e
GENERAL DISCUSSION
Two studies were conducted to examine the tenability of a model delineating
how individual differences in autonomy orientation and contextual autonomy sup-
port influence engagement in prosocial activities. Self-determination theory was
usedas the backbone ofthe model,and proposes thatthe satisfaction ofpsycholog-
ical needs for competence, autonomy, and relatedness mediates the effect of these
predictors on prosocial behavior. The first study used a cross-sectional design in
which college students reported on their past and current engagement in proso-
cial behavior, and on other variables in the model. Results from a path analysis
supported the hypothesized model, but showed that autonomy orientation was a
stronger predictor of engagement than autonomy support was. Need satisfaction
mediated the relation between these two predictors and engagement in prosocial
activities.
The second study employed a prospective design where volunteers from an
animal shelter reported on the autonomy supportiveness of the work environ-
ment, and their need satisfaction at work. They were also asked to report on
their psychological engagement in volunteer activities, and the number of hours
they volunteered during the 4-month period following the initial questionnaire.
Results partially supported the model. Autonomy orientation directly influenced
psychologicalengagement,whileworkautonomy support marginally predicted the
number of volunteered hours. Deci et al. (2001) found that autonomy support was
relatedtopsychological engagement for paid workers from two different countries.
Although these results were not replicated in the current study with volunteers, the
number of volunteered hours was nonetheless affected by autonomy support. The
number of volunteered hours is a good indication of how dedicated the volunteers
were, as it is assumed that volunteers work because they want to, not because they
have to. This fact, that people volunteer because they want to, not because external
reasons push them to, may explain why autonomy orientation, an individual differ-
ence variable, may have overridden the effects of contextual autonomy support, as
indicated by its stronger relation to engagement. Need satisfaction was also found
to mediate the relation between autonomy support and quantitative engagement,
but mediation between autonomy orientation and psychological engagement was
not supported.
Finally, autonomy support predicted volunteer turnover, which is in line with
previous results in education (Vallerand et al., 1997) and organizations (Rhoades
et al., 2001). Even though volunteers in this study were drawn from the same
workplace,theyreported experiencing different levels of autonomy support.These
differences may be caused by the fact that different volunteers tend to work during
a particular time of the week, which may be more or less busy, work on different
tasks, and with different staff. Work volume, the nature of the tasks, and the
autonomysupportivenessofparticular staff could all differentiallyaffectthe extent
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 219
to which volunteers perceived their work environment to be autonomy supportive.
Information about the nature of the tasks most often completed by volunteers
(e.g., cleaning cages, walking dogs, greeting visitors, conducting pet therapy, fund
raising by phone) was examined, but did not predict retention. Other detailed
information,suchasthe particular staff members with whom the volunteer worked,
was not available. Thus, although all volunteers worked at the same shelter and
thus within the same overall environment, there may have been other features of
the task, personality factors, or differences in expectations and preferences that
lead them to perceive a particular aspect of their volunteer experience differently.
The Role of Internalization in Motivating Prosocial Behavior
Theresultsofthepresentstudiesareinagreementwithpreviousstudiesexam-
iningthe effects of inductive and authoritative parenting on self-regulation (Grusec
&Goodnow,1994). But how does this process function?Self-determination theory
proposes that there is an internalization process (Deci & Ryan, 1991; Ryan, 1995)
that is facilitated by autonomy support. As an extension to the current studies, fu-
tureresearch could examine how internalization wouldpromote the self-regulation
of prosocial behavior. It would also be interesting to examine the role of auton-
omy support in the development of empathy, which has been shown to influence
engagement in prosocial behavior (Eisenberg, Miller, Shell, McNalley, & Shea,
1991). An interesting question would be whether prosocial behavior is a function
of internalization of the value of such behavior, or a direct outcome of need sat-
isfaction. Some research has found a direct link between need satisfaction and
community values, controlling for the influence of internalization of parental val-
ues (Kasser et al., 1995). But it is also likely that internalization also serves as the
basis through which prosocial behavior is self-regulated and maintained.
Implications for Promoting Prosocial Behavior
Earlier, I discussed the effects of controlling environments (e.g., mandatory
volunteering programs) on volitional engagement in prosocial behavior, and the
results of the present studies suggest that making volunteering programs a re-
quirement for high school graduation might not be the best means for promoting
self-regulated prosocial behavior. These findings, coupled with findings on induc-
tive parenting, suggest instead that using autonomy support would be more a more
effective means of fostering the self-regulation of prosocial activities in young
members of our societies. But the results of the present studies extend beyond
volunteer behavior. They also have interesting implications for organizational be-
havior.Penner,Midili, and Kegelmeyer(1997) argued that psychological processes
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
220 Gagn´e
involved in the motivation of volunteer behavior are applicable prosocial behav-
iors in organizations, which researchers call organizational citizenship behaviors
(Organ, 1988). Future research could therefore examine the effects of autonomy
support and need satisfaction on engagement in organizational citizenship behav-
ior. If such research shows that autonomy support positively affects engagement in
organizational citizenship behavior, implications for management will be to train
managers to act in autonomy supportive ways with subordinates. Such training has
been used before and has been found to have effects on the job satisfaction and
organizational trust of subordinates (Deci et al., 1989).
Future Research
A limitation of this study is the possible problem of self-selection, whereby
those who completed the questionnaires may have been more motivated than those
whodidnot.Thisisreflectedintheanalysiscomparingthenumberofhoursworked
by participants versus nonparticipants. Another limitation is the reliance on self-
reports for measuring autonomy support. This measure essentially represents only
perceived autonomy support. Future studies should try to obtain reports from
other sources, such as parents and managers, or use observational methods to
examine how supportive the environment is. Data from multiple sources would
provideadditionalvaliditytothis model. Moreover,one wouldneed to examine the
causal relation between autonomy support and increased engagement in prosocial
behavior, perhaps through laboratory studies or well-crafted quasi-experimental
field studies. For now, we can rely on evidence from past laboratory studies, such
as the one by Fabes et al. (1989) and the one by Deci et al. (1994), which found
support for a causal relation between autonomy support and increased behavioral
engagement.
Despite these limitations, the current studies demonstrate that self-
determination theory can be applied to the domain of prosocial motivation, and
suggest new avenues for the study of this motivation. What the present results
tentatively suggest is that if we want people to act prosocially, social structures
that fulfill basic psychological needs should be encouraged. People who feel com-
petent, volitional, and related to their peers may be more likely to be motivated to
engage in behaviors that are more prosocial.
REFERENCES
Allison, P. D. (1999). Multiple regression: A primer. Thousand Oaks, CA: Pine Forge Press.
Amabile, T. M. (1979). Effects of external evaluations on artistic creativity. Journal of Personality and
Social Psychology,37, 221–233.
Amabile, T. M., DeJonc, W., & Lepper, M. R. (1976). Effects of externally imposed deadlines on
subsequent intrinsic motivation. Journal of Personality and Social Psychology,34, 92–98.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 221
Arbuckle, J. L. (1999). Amos (Version 4) [Computer software]. Chicago, IL: Small Waters Corp.
Baard, P. P., Deci, E. L., & Ryan, R. M. (in press). The relation of intrinsic need satisfaction to
performance and well-being in two work settings. Journal of Applied Social Psychology.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psycho-
logical research: Conceptual, strategic and statistical considerations. Journal of Personality and
Social Psychology,51, 1173–1182.
Batson, C. D., Coke, J. S., Jasnoski, M. L., & Hanson, M. (1978). Buying kindness: Effect of an
extrinsic incentive for helping on perceived altruism. Personality and Social Psychology Bulletin,
4, 86–91.
Batson, C. D., Sager, K., Garst, E., Kang, M., Rubchinsky, K., & Dawson, K. (1997). Is empathy-
induced helping due to self-other merging? Journal of Personality and Social Psychology,73,
495–509.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
Cialdini, R. B., & Kenrick, D. T. (1976). Altruism as hedonism: A social development perspective on
therelationship of negative moodstate and helping. Journal of Personalityand Social Psychology,
63, 30–40.
Clary,E.G., Snyder,M., Ridge, R. D., Copeland, J.,Stukas, A. A.,Haugen, J., etal. (1998). Understand-
ing and assessing the motivations of volunteers: A functional approach. Journal of Personality
and Social Psychology,74, 1516–1530.
deCharms, R. 1986. Personal causation: The internal affective determinants of behavior. New York:
Academic Press.
Deci,E. L.(1971). Effectsof externallymediated rewardsonintrinsic motivation.Journalof Personality
and Social Psychology,18, 105–115.
Deci, E. L., Connell, J. P., & Ryan, R. M. (1989). Self-determination in a work organization. Journal
of Applied Psychology,74, 580–590.
Deci, E. L., Eghrari, H., Patrick, B. C., & Leone, D. R. (1994). Facilitating internalization: The self-
determination theory perspective. Journal of Personality,62, 119–142.
Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examin-
ing the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin,125, 627–
668.
Deci, E. L., & Ryan, R. M. (1985a). Intrinsic motivation and self-determination in human behavior.
New York: Plenum.
Deci, E. L., & Ryan, R. M. (1985b). The General Causality Orientation Scale: Self-determination in
personality. Journal of Research in Personality,19, 119–134.
Deci, E. L., & Ryan, R. M. (1991). A motivational approach to self: Integration in personality. In
R. Dienstbier (Ed.), Nebraska symposium on motivation: Perspectives on motivation (Vol. 38,
pp. 237–288). Lincoln: University of Nebraska Press.
Deci, E. L., & Ryan, R. M. (2000). The “what” and the “why” of goal pursuits: Human needs and the
self-determination of behavior. Psychological Inquiry,11, 227–268.
Deci, E. L., Ryan, R. M., Gagne, M., Leone, D., Usunov, J., & Kornazheva, B. P. (2001). Need
satisfaction,motivation, andwell-being in the workorganizations of a formereastern bloc country.
Personality and Social Psychology Bulletin,27, 930–942.
Eisenberg-Berg, N. (1979). Development of children’s prosocial moral judgment. Developmental Psy-
chology,15, 128–137.
Eisenberg, N., Miller, P. A., Shell, R., McNalley, S., & Shea, C. (1991). Prosocial development in
adolescence: A longitudinal study. Developmental Psychology,27, 849–857.
Fabes, R. A., Fultz, J., Eisenberg, N., May-Plumlee, T., & Christopher, F. S. (1989). Effects of rewards
on children’s prosocial motivation: A socialization study. Developmental Psychology,25, 509–
515.
Frey, B. S. (1993). Motivation as a limit to pricing. Journal of Economic Psychology,14, 635–664.
Frone, M. R. (2000). Interpersonal conflict at work and psychological outcomes: Testing a model
among young workers. Journal of Occupational Health Psychology,5, 246–255.
Gagn´e, M.,Koestner,R., &Zuckerman, M.(2000). Facilitatingthe acceptance of organizational change:
The importance of self-determination. Journal of Applied Social Psychology,30, 1843–1852.
Grolnick, W. S., & Ryan, R. M. (1989). Parent styles associated with children’s self-regulation and
competence in school. Journal of Educational Psychology,81, 143–154.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
222 Gagn´e
Grolnick, W. S., Ryan, R. M., & Deci, E. L. (1991). The inner resources for school achievement: Moti-
vational mediators of children’s perceptions of their parents. Journal of Educational Psychology,
83, 508–517.
Grusec, J. E., & Goodnow, J. J. (1994). Impact of parental discipline methods on the child’s internal-
ization of values: A reconceptualization of current points of view. Developmental Psychology,30,
4–19.
Hackman, J. R., & Oldham, G. R. (1975). Development of the Job Diagnostic Survey. Journal of
Applied Psychology,60, 159–170.
Hoffman, M. L. (1984). Parent discipline, moral internalization, and development of prosocial motiva-
tion.In E. Staub,D. Bar-Tal,J. Karylowski, & J.Reykowski (Eds.),Development and maintenance
of prosocial behavior: International perspectives on positive morality (pp. 117–137). New York:
Plenum.
Hoyle, R. H., & Smith, G. T. (1994). Formulating clinical research hypotheses as structural equation
models: A conceptual overview. Journal of Consulting and Clinical Psychology,62, 429–440.
Ilardi, B. C., Leone, D., Kasser, T., & Ryan, R. M. (1993). Employee and supervisor ratings of motiva-
tion: Main effects and discrepancies associated with job satisfaction and adjustment in a factory
setting. Journal of Applied Social Psychology,23, 1789–1805.
Inagaki, K., & Hatano, G. (1984, April). Effects of external evaluation on reading comperehension and
intrinsicinterest. Paperpresented at themeeting of the AmericanEducation Research Association,
New Orleans.
Kasser, T., & Ryan, R. M. (1993). A dark side of the American Dream: Correlates of financial success
as a central life aspiration. Journal of Personality and Social Psychology,65, 410–422.
Kasser, T., Ryan, R. M., Zax, M., & Sameroff, A. J. (1995). The relations of maternal and social
environments to late adolescents’ materialistic and prosocial values. Developmental Psychology,
31, 907–914.
Kasser, T., & Sheldon, K. M. (2000). Of wealth and death: Materialism, mortality salience, and con-
sumption behavior. Psychological Science,11, 348–351.
Koestner, R., Franz, C., & Weinberger, J. (1990). The family origins of empathic concern: A 26-year
longitudinal study. Journal of Personality and Social Psychology,58, 709–717.
Koestner, R., Ryan, R. M., Bernieri, F., & Holt, K. (1984). Setting limits on children’s behavior: The
differential effects of controlling vs. informational styles on intrinsic motivation and creativity.
Journal of Personality,52, 231–248.
Kunda, Z., & Schwartz, S. H. (1983). Undermining intrinsic moral motivation: External reward and
self-presentation. Journal of Personality and Social Psychology,45, 763–771.
Latan´e, B., & Darley, J. M. (1970). The unresponsive bystander: Why doesn’t he help? New York:
Appleton-Century-Crofts.
Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Lexington,
MA: Lexington Books.
Paulhus, D. L., Shaffer, D. R., & Downing, L. L. (1977). Effects of making blood donor motives salient
upon donor retention: A field experiment. Personality and Social Psychology Bulletin,3, 99–102.
Penner, L. A., & Finkelstein, M. A. (1998). Dispositional and structural determinants of volunteerism.
Journal of Personality and Social Psychology,74, 525–537.
Penner, L. A., Midili, A. R., & Kegelmeyer, J. (1997). Beyond job attitudes: A personality and social
psychologyperspective onthe causes of organizationalcitizenship behavior.Human Performance,
10, 111–131.
Rhoades, L., Eisenberger, R., & Armeli, S. (2001). Affective commitment to the organization: The
contribution of perceived organizational support. Journal of Applied Psychology,86, 825–
836.
Robbins, R. J. (1994). An assessment of perceived parental autonomy support and control: Child and
parent correlates. Unpublished doctoral dissertation, University of Rochester.
Rosenthal, R., & Rosnow, R. L. (1984). Essentials of behavioral research: Methods and data analysis.
New York: McGraw-Hill.
Ryan, R. M. (1995). Psychological needs and the facilitation of integrative processes. Journal of
Personality,63, 397–427.
Ryan,R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation,
social development and well being. American Psychologist,55, 68–78.
P1: FLT
Motivation and Emotion [me] PP942-MOEM-470701 July 28, 2003 13:33 Style file version Nov 28th, 2002
Need Satisfaction and Prosocial Behavior 223
Ryan, R. M., Stiller, J., & Lynch, J. H. (1994). Representations of relationships to teachers, parents,
and friends as predictors of academic motivation and self-esteem. Journal of Early Adolescence,
14, 226–249.
Skinner, B. F. (1978). The ethics of helping people. In L. Wisp´e (Ed.), Altruism, sympathy, and helping:
Psychological and sociological principles (pp. 249–262). New York: Academic Press.
Sobus, M. S. (1995). Mandating community service: Psychological implications of requiring prosocial
behavior. Law and Psychology Review,19, 153–182.
Staub, E. (1978). Positive social behavior and morality: Vol. 1. Social and personal influences.New
York: Academic Press.
Stukas,A. A., Snyder,M., & Clary, E.G. (1999). The effectsof “mandatory volunteerism”on intentions
to volunteer. Psychological Science,10, 59–64.
Upton, W. E., III (1974). Altruism, attribution and intrinsic motivation in the recruitment of blood
donors. In Selected readings in donor recruitment (Vol. 2, pp. 7–38). Washington, DC: American
National Red Cross.
Vallerand, R. J. (1997). Toward a hierarchical model of intrinsic and extrinsic motivation. In M. P.
Zanna (Ed.), Advances in experimental social psychology (Vol. 29, pp. 271–360). San Diego:
Academic Press.
Vallerand, R. J., Fortier, M. S., & Guay, F. (1997). Self-determination and persistence in a real-life
setting: Toward a motivational model of high school dropout. Journal of Personality and Social
Psychology,72, 1161–1176.
Wang, J., Fisher, J. H., Siegal, H. A., Falck, R. S., & Carlson, R. G. (1995). Influence of measurement
errors on HIV risk behavior analysis: A case study examining condom use among drug users.
Structural Equation Modeling,2, 319–334.
Williams, G. C., & Deci, E. L. (1996). Internalization of biopsychosocial values by medical students:
A test of self-determination theory. Journal of Personality and Social Psychology,70, 767–779.
Williams, G. C., Freedman, Z., & Deci, E. L. (1998). Supporting autonomy to motivate patients with
diabetes for glucose control. Diabetes Care,21, 1644–1651.
Williams, G. C., Gagn´e, M., Ryan, R. M., & Deci, E. L. (2002). Facilitating autonomous motivation
for smoking cessation. Health Psychology,21, 40–50.
Williams, G. C., Grow, V. M., Freedman, Z., Ryan, R. M., & Deci, E. L. (1996). Motivational predictors
of weight loss and weight-loss maintenance. Journal of Personality and Social Psychology,70,
115–126.
Zuckerman, M., Porac, J., Lathin, D., Smith, R., & Deci, E. L. (1978). On the importance of self-
determination for intrinsically motivated behavior. Personality and Social Psychology Bulletin,4,
443–446.
