تاثیر انگیزش خود تعیین شده بر روی هویت نقش داوطلب: مطالعه پانل-عقب افتاده متقابل

Volunteer role identity is regarded the direct and proximal cause of sustained volunteerism. Self-determination theory suggests that the quality of motivation greatly affects performance and well-being in various contexts. Therefore, this study investigated cross-lagged effects (over a time period of 16 months) between self-determined and controlled motivation, on the one hand, and two types of volunteer role identities, on the other hand: general role identity (GRI) and organization-specific role identity (OSRI). Analyses confirmed positive time-lagged effects of self-determined motivation on both GRI and OSRI. Time-lagged effects in opposite direction were significantly weaker; only OSRI showed a positive impact on subsequent self-determined motivation. OSRI (but not GRI) also had a positive lagged effect on controlled motivation. Overall, this study supported the idea that self-determined motivation represents a causal antecedent of volunteer role identities.

Volunteering is an area of growing political and scientific interest. Many social psychologists addressed the motivational basis of this planned and sustained type of prosocial behavior. Several functions that can be served by volunteering have been identified, ranging from the expression of values to career benefits (cf. Clary et al., 1998). Integrating this so-called functional approach and role identity theory, Penner (2002) suggested that over the course of volunteering, people develop a volunteer role identity which becomes the proximal determinant of sustained volunteering. Making use of self-determination theory (SDT; cf. Deci and Ryan, 1985 and Deci and Ryan, 2000), the present study investigated the causality between the quality of volunteer motivation and volunteer role identity across time. SDT suggests that self-determined types of motivation are positively associated with a wide range of personal and organizational outcomes (cf., with respect to the work context, Gagné & Deci, 2005). In particular, self-determined motivation has been linked to affective organizational commitment—that is, to a variable that shares similarities with the construct of role identity (cf. Gagné et al., 2014). Given the discretionary nature of volunteering, we proposed that self-determination captures an important aspect of the volunteer experience affecting volunteer role identity. In a cross-sectional study, Finkelstein (2009) demonstrated that a trait-like orientation towards intrinsic motivation, which is one type of self-determined motivation, positively correlated with volunteer role identity. In the present study, we examined the link between the context-specific quality of motivation and two types of volunteer role identities. To address the causality of these relationships, we chose a cross-lagged panel approach using SEM. In field settings, this approach is more appropriate than any cross-sectional design to investigate the directionality of effects and to infer causality (cf. Zapf, Dormann, & Frese, 1996). 1.1. Volunteer role identities Role identity theory (cf. Grube and Piliavin, 2000 and Piliavin et al., 2002) underscores the social context of volunteerism. A volunteer role identity “concerns the extent to which a person identifies with and internalizes the role of being a volunteer; that is, the extent to which this role and the relationships associated with it become part of a person’s self-concept” (Penner, 2002, p. 463). The dimension of time is taken into account by Penner’s integrative framework. The initial decision to volunteer reflects demographical and dispositional aspects such as motives. Over the course of volunteering, organizational factors foster the development of a volunteer role identity, which is considered the direct and proximal cause of sustained volunteering. Grube and Piliavin (2000) differentiated between a general role identity (GRI) and an organization-specific role identity (OSRI) and demonstrated that different role identities may conflict. For volunteers with a strong GRI, being a volunteer means more than engaging in some prosocial behavior; volunteering represents a significant component of this person’s self-concept. Whereas GRI may relate to different kinds of volunteer activities, OSRI is linked to a particular setting for volunteering. A strong OSRI means that the role a volunteer plays in a specific organizational context and the relationships associated with this particular involvement have become an important part of who this person is. Volunteers with strong GRI, but comparatively week OSRI may quit volunteering for a specific organization in order to volunteer for another organization in a related field. Fostering OSRI represents a major challenge to non-profit organizations—particularly in a competitive environment. In this study, we explored whether GRI and OSRI were similarly affected by volunteers’ motivation or showed differential causal relationships. 1.2. Self-determination theory As a theory of human motivation and personality, SDT focuses on the quality of motivation, beyond a quantitative approach. SDT differentiates between types of motivation that are associated with either the experience of choice, volition, and self-determination or the experience of control, pressure, and alienation. Intrinsic motivation represents the prototype of self-determined motivation; it is characterized as the innate tendency to exercise one’s capacities, and to seek optimal challenges. Doing something out of intrinsic motivation means that the person is motivated by interest and enjoyment of the activity itself. In contrast, extrinsic motivation means that the respective activity is instrumental to a goal that can be separated from the activity itself. Extrinsic motivation refers to activities that are not inherently interesting and challenging. However, extrinsic motivation does not per se represent a controlled type of motivation. Even unpleasant extrinsically motivated behaviors can be regulated in a self-determined way. Different types of extrinsic motivation can be ordered according to their degree of self-determined regulation. External regulation implies that the behavior is regulated by external contingencies such as rewards or punishments. Introjected regulation implies that feelings of contingent self-worth, guilt, or shame are the rewards or punishments that regulate the behavior. Both external regulation and introjected regulation represent controlled motivation; the behavior is under the control of regulators that are external to the core sense of self. If, however, the extrinsically motivated behavior is regulated by personally important goals and values, this identified regulation is associated with the experience of self-determination. Both intrinsic motivation and identified regulation of extrinsic motivation represent self-determined motivation. Taken together, self-determined motivation means that one’s behavior is regulated by inherent interest or by goals and values people can identify with. Self-determined motivation is accompanied by the experience of volition and choice, and reflects human flourishing. Although some studies use composite scores of “relative self-determination”, we analyzed self-determined motivation and controlled motivation separately. Previous studies have shown differential relationships with other variables for self-determined and controlled motivation (cf. Bidee et al., 2013). The experience of self-determination is associated with widespread advantages concerning affective, behavioral, and cognitive outcomes. In the context of work and organization, self-determination was positively associated with several favorable work outcomes (cf. Gagné et al., 2014). In particular, self-determined motivation was linked to variables that are related to the concept of role identity such as occupational or organizational commitment. In a study with school principals, intrinsic motivation and identified regulation were strongly related to occupational commitment (Fernet, 2011). In a cross-lagged panel study by Gagné, Chemolli, Forest, and Koestner (2008), self-determined motivation showed a positive lagged impact on affective commitment, the opposite effect not being significant. The present study addressed the context-specific quality of volunteer motivation instead of trait-like tendencies examined by Finkelstein (2009). More importantly, this study investigated self-determined motivation instead of only intrinsic motivation. Intrinsic motivation (i.e., having fun, being challenged, etc.) captures an important, but only one aspect of the experience of self-determination; identified regulation (i.e., extrinsic motivation which is regulated by important goals and values) is equally important. The experience of doing something personally meaningful (which is not necessarily fun!) represents an essential facet of self-determined motivation—particularly, in the context of volunteering. 1.3. Hypotheses Against the background of SDT, we hypothesized that self-determined motivation allows volunteers to act in congruence with their core sense of self and, thus, qualifies as a causal antecedent of volunteer role identities. Penner (2002) posited that a “high and involving level of volunteer activity” (p. 463) fosters role identity. The concept of self-determined motivation fits to this description of high involvement. For cultural demands, roles, and values to be integrated into one’s self, the individual must have the opportunity to “freely process and endorse transmitted values and regulations” (Deci & Ryan, 2000, p. 238). We proposed that, for a role to become and remain an important part of the self-concept, it is essential that the respective behaviors are regulated in a self-determined way. A behavior that is neither personally meaningful nor experienced as intrinsically motivating is very unlikely to constitute an integral part of the self. On the contrary, if these behaviors reflect inherent interest or important values, the corresponding roles are expected to become aspects of who a person is. Thus, we hypothesized: Hypothesis 1. Self-determined motivation has a positive cross-lagged effect on (a) GRI and (b) OSRI. Taking a longitudinal perspective, we proposed self-determined motivation as an antecedent rather than a consequence of volunteer role identities. We did not reject the possibility of cross-lagged effects in opposite direction (cf. Meyer, Becker, & Vandenberghe, 2004, who suggested short-term effects of commitment on motivation and long-term effects in reverse direction). However, in line with the Gagné et al. (2008) study on affective commitment, we expected the positive impact of self-determined motivation on role identities to be stronger than the reverse effects. Hypothesis 2. The positive cross-lagged effects of self-determined motivation on (a) GRI and (b) OSRI are stronger than the corresponding cross-lagged effects in opposite direction. We made no specific predictions about cross-lagged effects of controlled motivation on volunteer role identities. Compared to self-determined motivation, controlled motivation has often shown considerably weaker and sometimes inconsistent relationships with work outcomes (cf. Gagné et al., 2014). Thus, we did not hypothesize substantial lagged effects of controlled motivation on either GRI or OSRI, but only explored these relationships from a longitudinal perspective.

3. Results 3.1. Descriptive statistics and measurement invariance Means, standard deviations, intercorrelations, and internal consistencies for all variables are shown in Table 1. At both T1 and T2, self-determined motivation and both types of role identity were positively correlated. Controlled motivation showed weaker, but significant positive correlations with GRI and OSRI at both times. Both GRI (t = 9.78, p < .001) and OSRI (t = 9.77, p < .001) increased from T1 to T2. Measurement invariance across time was fully supported. For all study variables, the free loadings models showed adequate fit to the data, and the fit of the respective invariant loadings models was not significantly poorer, supporting the notion that the indicator variable measure the same latent constructs across time. 3.2. Structural cross-lagged relationships The cross-lagged relationships of all four variables were concurrently addressed. This overall model was based on the assumption of measurement invariance across time, which had been supported before. The overall model provided adequate fit to the data: χ2 = 402.57, df = 208, p < .001; χ2/df ratio = 1.94; CFI = .987; RMSEA = .034; SRMR = .028. Estimates for stabilities and cross-lagged effects are shown in Table 2. Hypothesis 1 was supported: Self-determined motivation at T1 had a positive lagged effect on both GRI and OSRI at T2. In the reverse direction, a positive lagged effect was found for OSRI, but not for GRI. The lagged effects of self-determined motivation on GRI and OSRI were stronger than the corresponding reversed effects. To test Hypothesis 2, we compared the structural model as a baseline model to a second model with equality constraints. In this second model, the lagged effects of self-determined motivation on GRI and OSRI were set equal to the corresponding reverse effects of GRI and OSRI on self-determined motivation. Adding the equality constraints significantly decreased model fit (Δχ2 = 22.13, df = 2, p < .001). Taken together, Hypothesis 2 was also supported. Table 2. Standardized estimates for the structural relationships. Effect Estimate 95% CI Lower limit Upper limit Stabilities Self-determined motivation (SD) .689⁎⁎⁎ .593 .767 Controlled motivation (CO) .720⁎⁎⁎ .661 .774 General role identity (GRI) .493⁎⁎⁎ .384 .600 Organization-specific role identity (OSRI) .462⁎⁎⁎ .350 .572 Cross-lagged effects: SD and role identities SD → GRI .229⁎⁎⁎ .119 .333 SD → OSRI .191⁎⁎ .082 .299 GRI → SD .024 −.061 .112 OSRI → SD .114⁎ .023 .207 Cross-lagged effects: CO and role identities CO → GRI .004 −.071 .081 CO → OSRI .068 −.007 .144 GRI → CO .052 −.037 .135 OSRI → CO .101⁎ .016 .186 Other cross-lagged effects SD → CO −.194⁎⁎⁎ −.280 −.112 CO → SD −.051 −.121 .021 GRI → OSRI .103 −.015 .217 OSRI → GRI .001 −.092 .096 Note. N = 805. CI = confidence interval, bias-corrected percentile method, based on k = 5’000 bootstrap samples. ⁎ p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001, two-tailed. Table options Both GRI and OSRI increased over time. Thus, cross-lagged effects might be limited to volunteers scoring low on GRI/OSRI at T1. Exploration, however, revealed that T1 self-determined motivation showed comparably strong positive associations with residuals of T2 GRI/OSRI regressed on T1 GRI/OSRI for participants at all baseline levels of GRI/OSRI. None of the hypothesized effects changed when we controlled for constant background variables (e.g., age, gender). To gauge potential biases due to nonconstant third variables, we contrasted the fit of cross-lagged versus common factor models (cf. Link & Shrout, 1992). For each pair of variables (e.g., self-determined motivation and GRI), the common factor model fit was considerably weaker than the cross-lagged model fit, indicating that the cross-lagged models were not caused by a nonconstant third variable. We explored the cross-lagged associations between controlled motivation and volunteer role identities. Controlled motivation at T1 had no impact on either GRI or OSRI at T2. However, the lagged effect of OSRI (but not GRI) on controlled motivation was positive. Finally, self-determined motivation showed a negative lagged effect on controlled motivation.