The Impact of Leadership Style on Volunteer Motivation

Abstract 

This study examines the relationship of alternative leadership styles exercised by senior church leaders with the motivation of church volunteers. Leadership that may impact the motivation of members to fulfil volunteer roles in church settings has received very limited attention within empirical leadership and organizational behavior research. However, maintaining volunteer motivation at levels that result in sustained and productive service is critical to the effectiveness of socially important non-profit organizations such as churches in fulfilling their stated missions. The study contributes to the literature by applying key organizational leadership theories to alternative conceptualizations of the nature of motivation of unpaid volunteers and empirically examining the resulting relationships. 

This study investigated four theory based hypotheses exploring relationships between transactional and transformational leadership behaviours and volunteer extrinsic and intrinsic motivation with trust and value congruence as mediating variables. The findings are based on responses from a sample of 790 persons who served in a voluntary capacity within their congregation and who rated the leadership behaviours of their senior pastor. These volunteers attended and participated in 28 different Australian Christian Churches (ACC) congregations in Australia. Multivariate analyses showed that three of the four hypotheses were supported. Transactional leadership predicted extrinsic motivation, but transformational leadership did not. Transformational leadership predicted intrinsic motivation, but transactional leadership did not. Trust and value congruence both exerted partial mediation on the relation between transformational leadership and intrinsic motivation. Theoretical and practical implications of these results are presented and suggestions for future research are discussed. 


Introduction 

Leaders of non-profit organisations often rely upon their ability to articulate a value-based vision and to model appropriate behaviours to motivate volunteers (Druskat, 1994; Onnen, 1987; Rowold, 2008; Rowold and Rohmann, 2009). How can such leaders motivate volunteers to higher levels of commitment and performance? Although this is a question of theoretical and practical importance, relatively few studies have explored organisational leadership style and outcomes within non-profit organisations (Bae, 2001; Balswick and Wright, 1988; Butler and Herman, 1999; Callahan, 1996; Catano, et al., 2001; Choi, 2006; Crain-Gully, 2003; Druskat, 1994; Johnson, 2007; Knudsen, 2006; Larsson and Ronnmark, 1996; Lichtman and Malony, 1990; Onnen, 1987; Rowold, 2008; Rowold and Rohmann, 2009; Son, 2003). The few previous studies have not focused on the specific relationship between leadership style and volunteer motivation. The present study contributes to the literature by examining the relationship of a senior leader’s leadership style with volunteer motivation in non-profit organisations. The study focuses specifically on transactional and transformational leadership behaviours. In addition, follower trust in and value congruence with a leader are investigated as possible mediators in the leader-volunteer relationship (Jung and Avolio, 2000; Podsakoff, et al., 1990; Shamir, et al., 1993). 


Theoretical Background 

Volunteer Motivation 

Volunteers are individuals who provide unpaid help in an organised manner to parties with regard to whom the volunteer has no obligations (Millette and Gagné, 2008; Snyder and Omoto, 2004; Wilson and Janoski, 1995). Volunteers are eagerly sought after by non-profit entities such as churches and charities because they add value to organisations and their endeavours (Wilson and Musick, 1997; Phillips and Phillips, 2010, 2011). Because volunteers do not receive direct personal tangible gains such as a salary, non-profit organisations must find other ways to motivate them to work well and to continue in volunteer activity, and by doing so retain the knowledge and skill resources of the organisation (Millette and Gagné, 2008). Maintaining volunteer motivation at levels that result in sustained and productive voluntary service is critical to the effectiveness of non-profit organisations in fulfilling their stated missions. 

Volunteer motivation can be conceptualised using self-determination theory (Deci and Ryan, 2000, 2008), which posits that people are motivated to satisfy their innate psychological needs for autonomy, competence, and relatedness. Autonomy refers to the desire to control one’s own behaviour and activities in order to experience personal integration and freedom. Competence is one’s propensity to be effective in dealing with the environment while attaining valued outcomes within it. Relatedness refers to one’s desire to be connected to others. 

The satisfaction of all three of these needs is related to levels of volunteer extrinsic and intrinsic motivation (Deci and Ryan, 2000). Extrinsic motivation refers to volunteering for an activity for instrumental reasons, such as acquiring a reward or avoiding a penalty, where the primary motivators are external to the volunteer. By contrast, intrinsic motivation refers to volunteering for an activity for its own sake, because one finds it enjoyable and interesting, where the primary motivators are internal to the volunteer as s/he seeks to fulfil the needs for autonomy, competence, and relatedness. Extrinsic motivation has been demonstrated to predict lower quality task performance and shorter volunteer tenure whereas intrinsic motivation predicts higher quality task performance and longer volunteer tenure (Deci and Ryan, 2008; Millette and Gagné, 2008). 

Transactional and Transformational Leadership 

Transactional and transformational leadership behaviours have been identified as appropriate and effective components of supervisory style within non-profit organisations (Bae; 2001; Balswick and Wright, 1988; Butler and Herman, 1999; Callahan, 1996; Catano et al., 2001; Choi, 2006; Crain-Gully, 2003; Druskat, 1994; Johnson, 2007; Knudsen, 2006; Larsson and Ronnmark, 1996; Onnen, 1987; Rowold, 2008; Rowold and Rohmann, 2009; Son, 2003). 

Transactional leadership involves a reciprocal process of exchange between leader and followers (Bass, 1985; Riggio et al., 2004). It is defined in terms of three inter-related behaviours: (a) contingent reward, (b) active management by exception, and (c) passive management by exception (Yammarino and Bass, 1990). Contingent reward implies the provision of an adequate exchange of valued resources for follower support (Judge and Bono, 2000). Active management by exception involves monitoring performance and taking corrective action. Passive management by exception means intervening only when problems become serious. Both active and passive management by exception involve enforcing rules to avoid mistakes (Judge and Bono, 2000). Contingent reward behaviours provide volunteers with a clear understanding of their tasks and the desired outcomes, and create an expectation of rewards for achievement, both of which assist volunteers in improving their performance. 

Transformational leadership is built around the notion that leaders and followers are held together by “some higher-level, shared goal or mission, rather than because of some personal transaction” (Riggio et al., 2004, p. 51). It involves a reciprocal process of inspiration between leader and followers which results in both performing beyond expected levels of commitment and contribution, and which is based on the leader developing “positive, rich, emotional relationships with followers that build commitment to a common purpose or cause” (Riggio et al., 2004, p. 50). Transformational leadership is defined in terms of four interrelated behaviours: (a) idealised influence, (b) inspirational motivation, (c) intellectual stimulation, and (4) individualised consideration (Riggio et al., 2004; Yammarino and Bass, 1990). Idealised influence involves leaders serving as idealised role models for followers (Avolio and Bass, 2004; Judge and Bono, 2000; Riggio et al., 2004). Inspirational motivation “arouses followers’ enthusiasm and sense of team spirit” (Riggio et al., 2004, p. 51) as the leader provides followers with a clear vision of the organisation’s future, the value of high standards of operation, and a sense of meaningfulness in their work (Avolio and Bass, 2004). Intellectual stimulation involves leaders encouraging followers to be innovators and creative problem solvers (Avolio and Bass, 2004; Yammarino and Bass, 1990). Individualised consideration involves the leader’s attention to the unique gifts and talents of each follower and the leader’s ability to coach or mentor followers with challenges and opportunities that suit each individual (Avolio, et al., 1999; Bass, 1985; Yammarino and Bass, 1990; Yammarino, et al., 1993). 

Overall, transformational leadership encapsulates a “sense of moral good and a passionate commitment to the cause” (Riggio et al., 2004, p. 52) that is essential for leadership in non-profit organisations which are mission-driven and which rely on the motivation and performance of volunteers to achieve the organisational mission. The impact of transformational leadership behaviours on volunteer workers may be to augment the effect of transactional leadership behaviours by providing volunteers with vision and values to motivate them to continue in voluntary activity at high levels of volition and quality performance (Bass, 1985; Judge and Piccolo, 2004). 

Leadership Style and Volunteer Motivation 

Transactional leadership behaviours are likely to produce extrinsic motivation in volunteers as they are motivated to attain contingent rewards such as personal recognition or standing within the organisation. Transformational leadership behaviours are likely to produce intrinsic motivation as volunteers are motivated by identification with and commitment to the mission of the organisation. This personal identification and commitment is internally driven and volunteers are likely to sense that they are satisfying the needs of autonomy, competence, and relatedness. A leader’s exercise of transformational behaviours will augment the impact of his or her use of transactional behaviours by stimulating volunteer intrinsic motivation and producing more sustained and higher quality task performance. 

Trust and Value Congruence 

The effective exercise of leadership is based upon leader–follower relationships that incorporate followers’ trust in and value congruence with the leader (Yukl, 2006). Trust in a leader is “faith in and loyalty to the leader” (Podsakoff et al., 1990, p. 113). Value congruence with a leader is belief that the follower’s personal values are congruent with and aligned with those of the leader (Posner, 2010). Both transactional and transformational leadership behaviours can inspire trust and value congruence in followers. 

Transactional leaders may build followers’ trust by engaging in consistent behaviour and by honouring agreements (Bass, 1985; Jung and Avolio, 2000; Podsakoff, et al., 1990). They may stimulate followers’ value congruence by identifying mutual aspirations and acknowledging followers’ expertise, experience, and education (Jung and Avolio, 2000). Transactional leadership behaviours generate followers’ trust in and value congruence with the leader in relation to the nature of the organisation, the task and outcomes required for efficient organisational operation, and the attendant contingent rewards. Jung and Avolio referred to such trust and value congruence as being “conditional” because they are established “through a reliable execution of contracts and exchanges” (p. 952). The trust and value congruence inspired in followers by a leader’s exercise of transactional leadership behaviours suffices to ensure a cooperative working relationship and the successful completion of the necessary tasks. The practice of transactional behaviours by non-profit leaders is likely to establish and maintain volunteers’ trust in the competence and fairness of the leader and volunteers’ value congruence with the practices and desired outcomes of the leader. This trust and value congruence is likely to mediate the impact of the leader’s transactional leadership behaviours on volunteers’ extrinsic motivation. 

Transformational leaders may increase followers’ trust levels by developing their skills and confidence to perform tasks and assume responsibility, by providing support and encouragement when necessary in the face of obstacles, difficulties, and fatigue, and through their own role modelling of desirable behaviour and willingness to engage in sacrifice in order to achieve the organisational vision (Bass and Avolio, 1990; Yukl, 2006). They may influence followers to adopt and internalise the leader’s values and vision by providing and communicating a desirable vision and by raising followers’ level of awareness about the importance and value of desired outcomes (Avolio and Bass, 1988; Jung and Avolio, 2000). Jung and Avolio (2000) point out that “value congruence achieved through a value internalisation process and demonstrated trust in the leader,” are core mediating aspects of transformational leadership (p. 950). The practice of transformational leadership behaviours by non-profit leaders is likely to increase volunteers’ trust in the character and competence of the leader and to produce change in volunteers’ values and to increase their value congruence with the leader. This increased trust and value congruence is likely to mediate the impact of the leader’s transformational behaviours on volunteer intrinsic motivation. 

The model presented in Figure 1 emphasises the relationship between the transactional and transformational leadership behaviours of senior pastors and the extrinsic and intrinsic motivation of volunteers. The relationship between the independent variable and the dependent variable can be direct or mediated by volunteer trust in and/or value congruence with the leader. 

 
SGF-Fig1.png
 

Figure 1: A model of the relationships between transactional and transformational leadership behaviours and volunteer motivation, with volunteer trust in and value congruence with the leader as mediating variables. 

The relationships portrayed in Figure 1 form the basis of the study hypotheses. For example, leaders of non-profit organisations who practice transactional behaviours are likely to inspire volunteers to be satisfied with their leadership (Druskat, 1994) and to rate them as effective leaders (Butler and Herman, 1999). Transactional behaviours can motivate volunteers towards the achievement of desired organisational outcomes (Son, 2003). The practice of transactional behaviours is likely to result in volunteers clearly understanding their task descriptions and the desired outcomes for the organisation that result from the performance of these tasks. Volunteers will perform their tasks so as to achieve the expressed expectations of the leader and to receive the associated contingent rewards including personal recognition and standing within the organisation. Therefore, the practice of transactional behaviours by senior pastors is likely to reinforce the extrinsic motivation of volunteers. Thus: 

H1: The transactional leadership behaviours of senior pastors will be more strongly related to volunteer extrinsic motivation than will transformational leadership behaviours. 

The leader’s transactional leadership behaviours including clear specification of tasks, performance expectations, and contingent rewards builds follower trust in the leader (Bass, 1985; Jung and Avolio, 2000) and value congruence with the leader (Jung and Avolio, 2000). The practice of transactional behaviours is likely also to establish and maintain volunteer trust in the competence and fairness of the leader and volunteer value congruence with the practices and desired outcomes of the leader. Trust and value congruence in turn are likely to increase volunteer extrinsic motivation. Thus: 

H2: Volunteer trust in and value congruence with senior pastors will mediate the relationship of transactional leadership behaviours of senior pastors with volunteer extrinsic motivation. 

The transformational leadership behaviours of non-profit leaders are likely to inspire volunteers to be satisfied with their leadership (Druskat, 1994; Onnen, 1987; Rowold, 2008) and to motivate volunteers towards the achievement of desired organisational outcomes (Bae, 2001; Onnen, 1987; Son, 2003). These behaviours are also likely to produce increased volunteer morale and participation (Larsson and Ronnmark, 1996), volunteer job satisfaction (Choi, 2006; Rowold, 2008), and volunteer extra effort and effectiveness in task performance (Rowold, 2008; Rowold and Rohmann, 2009). 

Transformational leadership behaviours are likely to help volunteers understand and identify with the importance of the tasks they are undertaking and how these tasks contribute to the mission of the organisation. Volunteers will perform their tasks to a degree beyond the expressed expectations of the leader because they are motivated by the vision and values of the organisation, the personal meaningfulness of their involvement, and the sense of goal attainment and positive feelings that they experience. Thus: 

H3: The transformational leadership behaviours of senior pastors will be more strongly related to volunteer intrinsic motivation than will transactional leadership behaviours. 

Follower trust in and value congruence with a leader have been demonstrated to have a mediating effect on the relationship between transformational leadership behaviours and volunteer performance (Jung and Avolio, 2000; Podsakoff et al., 1990; Shamir, et al., 1993). The leader’s clear, attractive, and attainable vision for the organisation and demonstrated commitment to the organisation inspires follower trust in and value congruence with the leader and motivates them to do more than they originally expected to do (Bennis and Nanus, 1985; Avolio and Bass, 1995; Yukl, 2006). The practice of transformational behaviours is likely to increase volunteer trust in the character and competence of the leader and to produce change in volunteer values and to increase their value congruence with the leader. This increased trust and value congruence in turn may increase volunteer intrinsic motivation. Thus: 

H4: Volunteer trust in and value congruence with senior pastors will mediate the relationship of transformational leadership behaviours of senior pastors with volunteer intrinsic motivation. 

Control Variables 

It is likely that the age, gender, and length of tenure of participating volunteers could influence their assessment of their own motivation and of the leader’s style. It is also possible that the length of tenure of the senior pastor and congregation size could influence volunteer motivation. Therefore, the study incorporated the control variables of age, gender, and length of tenure of the volunteer participant, length of tenure of the senior pastor, and congregational size. 

Methodology 

Sample and Procedure 

The sample for this study was drawn from volunteers attending and participating in 28 different Australian Christian Churches (ACC) congregations in Australia. The sample consisted of 790 subjects who served in a voluntary capacity within their congregation and who rated the leadership behaviours of their senior pastor. The selection of 28 different ACC congregations was designed to provide responses from volunteers within each of five congregational size categories that ACC recognises. ACC, also known as the Assemblies of God in Australia, is a fellowship of autonomous churches which had 1,087 registered churches throughout Australia as of May 17, 2010 (ACC, 2010). The 28 congregations represent 2.6% of the total number of ACC congregations. ACC conducts an annual census of all congregations in May. The census collects data on various church activities, including weekend attendance which is measured as the total number of attendees at all services from Friday evening to Sunday evening on one weekend in May each year. ACC categorises congregational sizes into five categories: (a) under 100 attendees, (b) 100-199 attendees, (c) 200-499 attendees, (d) 500-999 attendees, and (c) 1000 and more attendees. In 2010, the average size of an ACC congregation was 208 attendees. 

An initial approach was made to the senior pastors of selected potential participating congregations explaining the study and inviting their participation. This was followed up with an email detailing the study. As affirmative responses were received, appointments were made for the researcher and/or research assistants to attend a Sunday worship service or a midweek leaders meeting in order to obtain responses from volunteers. During the worship service or leaders’ meeting, a short explanation of the study was made to those in attendance. Congregational members who served as volunteers in the church were invited to meet with the researcher and/or research assistants after the service or meeting in order to participate in the study. 

A Volunteer Response Pack was provided to each participating volunteer, containing: (a) a covering letter, (b) an informed consent form, (c) a personal demographic form, and (d) the questionnaire. The control variables of age, gender, and length of tenure of the subject were incorporated into the demographic form. The researcher and research assistants helped subjects to complete and submit the survey immediately after the service or meeting. 

A Senior Pastor Response Pack was provided to each participating senior pastor, containing: (a) a covering letter; (b) an informed consent form; (c) a personal demographic form; and (4) an adapted version of the questionnaire consisting of 32 items measuring their self-rating of their practice of transactional and transformational leadership behaviours. The control variable of senior pastor tenure was incorporated into and ascertained from the personal demographic form. 

Congregations were approached to participate in the study on the basis that they contributed to a representative sample of ACC congregational sizes. The 28 participating congregations together represented each of the five ACC congregational size categories. An average of just over 28 volunteers completed the survey in each of the 28 participating congregations resulting in a total participation of 790 volunteers. A summary of participating congregations and volunteers is displayed in Table 1. 

Table 1: Participating Congregations and Volunteers by Congregational Size 

 
SGF-Table1.png
 

The sample comprised of 365 males (46.2%) and 425 females (53.8%). Participating volunteers were well spread over five age categories (Table 2) and most had served as volunteers in their congregation for more than 5 years (Table 3). 

Table 2: Demographics—Volunteer Age Groups 

 
SGF-Table2.png
 

Table 3: Demographics—Volunteer Tenure 

 
 

Measures 

The survey questionnaire employed 56 items for measuring the variables. 

Volunteer Motivation 

Volunteers’ assessment of their own motivation to engage in voluntary activity was measured using an adapted form of the Volunteer Motivation Scale (VMS), developed by Millette and Gagné (2008), and consisting of 12 items loaded onto the two scales of extrinsic motivation and intrinsic motivation utilising a 5-point Likert scale: 0 (completely disagree), 1 (disagree), 2 (not sure), 3 (agree), 4 (completely agree). The reliabilities were α = .76 for extrinsic motivation and α = .74 for intrinsic motivation. 

Transactional and Transformational Leadership Behaviours 

Volunteers’ assessment of the senior pastor’s transactional and transformational leadership behaviours was measured using an adapted form of the MLQ-5X Rater Form consisting of 32 items loaded onto the two leadership scales 

(Avolio and Bass, 2004) utilising a 5-point Likert scale: 0 (not at all), 1 (once in a while), 2 (sometimes), 3 (fairly often), 4 (frequently, if not always). The reliabilities were α = .68 for transactional leadership and α = .84 for transformational leadership. 

Trust 

Volunteer trust in the senior pastor was measured using an adapted form of the Trust in the Leader Scale (TLS), developed by Podsakoff et al. (1990), consisting of six items loaded onto the one scale utilising a 5-point Likert scale: 0 (completely disagree), 1 (disagree), 2 (not sure), 3 (agree), 4 (completely agree). The reliability was α = .76. 

Value Congruence 

Volunteer value congruence with the senior pastor was measured using an adapted form of the Values Congruency Index (VCI), developed by Posner (1992, 2010), consisting of six items loaded onto the one scale utilising a 5-point Likert scale: 0 (completely disagree), 1 (disagree), 2 (not sure), 3 (agree), 4 (completely agree). The reliability was α = .78. 

Results 

The mean scores and standard deviations of the independent, dependent, mediating, and control variables and correlations among the variables are shown in Table 4. 

Table 4: Mean Scores, Standard Deviations, and Correlations among Study Variables (N = 790)

 
SGF-Table4.png
 

Tests of Hypotheses 

Hierarchical multivariate regression analyses were conducted to test the four hypotheses. Examination of Normal Probability Plots (P-P) and scatterplots confirmed the normality and homoscedasticity of the regression residuals for the relationships predicted by all four hypotheses. Examination of the Tolerance and Variance Inflation Factor (VIF) values for each variable demonstrated the absence of multicollinearity among the independent and control variables. Congregational size was dummy coded to provide a series of dichotomous variables in order to measure its impact on the relationships studied (Hair et al., 2010; Pallant, 2011). 

Hypothesis 1 

Hypothesis 1 was supported. The control variables were entered at Step 1, explaining 1.3% of variance in extrinsic motivation. After entry of transactional leadership and transformational leadership at Step 2, the total variance explained by the model as a whole was 7.7% (F [10, 779] = 6.48, p < .001). The two independent variables of transactional leadership and transformational leadership explained an additional 6.4% of the variance of extrinsic motivation (ΔR2 = .064, ΔF [2, 779] = 26.90, p < .001). In the final model, two variables were statistically significant, with transactional leadership recording the higher beta value (β = .26, p < .01) and volunteer age the lower beta value (β = -.08, p < .05). Transactional leadership predicted extrinsic motivation, but transformational leadership did not do so. The regression coefficients are presented in Table 5. 

Table 5: Regression Coefficients for H1 – Transactional Leader 

 
SGF-Table5.png
 


Hypothesis 2 

Hypothesis 2 was not supported. Trust and value congruence were regressed separately on transactional leadership to determine if it predicted either. In both cases there was no statistically significant effect of transactional leadership. If transactional leadership had been shown to have a statistically significant effect on trust or value congruence, then the following step would have been to test the mediation of trust and value congruence on the relationship between transactional leadership and extrinsic motivation. The regression coefficients are presented in Tables 6 and 7. 

Table 6: Regression Coefficients for H2 – Transactional Leadership Predicting Trust 

 
SGF-Table6.png
 


Table 7: Regression Coefficients for H2 – Transactional Leadership Predicting Value Congruence 

 
SGF-Table7.png
 

Hypothesis 3 

Hypothesis 3 was supported. The control variables were entered at Step 1, explaining 3% of variance in intrinsic motivation. After entry of transformational leadership and transactional leadership at Step 2, the total variance explained by the model as a whole was 13.3% (F [10, 779] = 11.99, p < .001). The two independent variables of transformational leadership and transactional leadership explained an additional 10.3% of the variance of IM after controlling for the other variables (ΔR2 = .103, ΔF [2, 779] = 46.37, p < .001). In the final model, five variables were statistically significant. Transformational leadership recorded the highest beta value (β = .32, p < .01). The other four statistically significant variables were congregational size 200-499 (β = .14, p < .01), congregational size ≥ 1000 (β = .13, p < .05), congregational size 500-999 (β = .12, p < .05), and volunteer gender (β = .08, p < .05). Transformational leadership predicted intrinsic motivation, but transactional leadership did not do so. The regression coefficients are presented in Table 8.

Table 8: Regression Coefficients for H3 – Transformational Leadership and Transactional Leadership Predicting Intrinsic Motivation 

 
SGF-Table8.png
 


Hypothesis 4 

Hypothesis 4 was supported. It was tested by conducting initial regression analyses to determine if transformational leadership predicted trust or value congruence, and then conducting further regression analyses to determine whether trust and value congruence had mediating effects on the relation between transformational leadership and intrinsic motivation. Control variables were not used in these regressions. 

Trust was regressed on transformational leadership to determine if transformational leadership predicted trust. This regression explained 43.4% of the variance in trust (ΔR2 = .434, ΔF [1, 778] = 604.39, p < .001). Transformational leadership was statistically significant, recording a beta value of .66, p < .01. Transformational leadership significantly predicted trust. The regression coefficients are presented in Table 9. 


Table 9: Regression Coefficient for H4 – Transformational Leadership Predicting Trust 

 
SGF-Table9.png
 

In order to test the mediating effect of trust on the relation between transformational leadership and intrinsic motivation further regression analysis was conducted. In a model predicting volunteer intrinsic motivation, transformational leadership was entered in Step 1, explaining 10.3% of variance in intrinsic motivation. After entry of trust in the leader in Step 2 of this model, the total variance explained by the model as a whole was 11.2% (F [2, 787] = 49.60, p < .001). In the final model, transformational leadership (β = .24, p < .01) and trust (β = .13, p < .01) were statistically significant. The change in the regression coefficient for transformational leadership indicates the mediation by trust. Transformational leadership had a regression coefficient of .43 (SE = .05, p < .01) at Step 1 and a regression coefficient of .32 (SE = .06, p < .01) at Step 2.

The statistical significance of the difference between the two transformational leadership regression coefficients (.43, .32) was tested by converting the values to a z score (Paternoster, Brame, Mazerolle, and Piquero, 1998). The result was z = .43 - .32 / √ .0462 + .0602 = 1.47, not significant. Because the regression coefficient for transformational leadership was smaller after entry of trust, but not significantly smaller, and because the betas for transformational leadership and trust were both statistically significant, trust exerted a partial mediation on the relation between transformational leadership and intrinsic motivation. The regression coefficients are presented in Table 10. 


Table 10: Regression Coefficient for H4 – Transformational Leadership and Trust Predicting Intrinsic Motivation 

 
SGF-Table10.png
 

In the second test of mediation, value congruence was regressed on transformational leadership to determine if transformational leadership predicted value congruence. This regression explained 37.4% of the variance in value congruence (ΔR2= .374, ΔF [1, 788] = 471.44, p < .001). Transformational leadership was statistically significant, recording a beta value of .61, p < .01. Transformational leadership significantly predicted value congruence. The regression coefficients are presented in Table 11.

Table 11: Regression Coefficients for H4—Transformational Leadership Predicting Value Congruence

 
SGF-Table11.png
 

In order to test the mediating effect of value congruence on the relation between transformational leadership and intrinsic motivation, further regression analysis was conducted. In a model predicting volunteer intrinsic motivation, transformational leadership was entered at Step 1, explaining 10.1% of variance in intrinsic motivation. After entry of value congruence at Step 2 of this model, the total variance explained by the model as a whole was 10.5% (F [2, 787] = 46.38, p < .001). In the final model, transformational leadership (β = .27, p < .01) and value congruence (β = .08, p < .05) were statistically significant. The change in the regression coefficient for transformational leadership indicates the mediation of value congruence. Transformational leadership had a regression coefficient of .43 (SE = .05, p < .01) at Step 1 and a regression coefficient of .36 (SE = .06, p < .01) at Step 2.

The statistical significance of the difference between the two transformational leadership regression coefficients (.43, .36) was tested by converting the values to a z score (Paternoster, Brame, Mazerolle, and Piquero, 1998). The result was z = .43 - .36 / √ .0462 + .0582 = .95, not significant. Because the regression coefficient for transformational leadership was smaller after entry of value congruence, but not significantly smaller, and because the betas for transformational leadership and value congruence were both statistically significant, value congruence exerted a partial mediation on the relation between transformational leadership and intrinsic motivation. The regression coefficients are presented in Table 12.


Table 12: Regression Coefficient for H4 – Transformational Leadership and Value Congruence Predicting Intrinsic Motivation 

 
SGF-Table12.png
 


Transformational leadership significantly predicted trust and value congruence. Trust and value congruence both exerted partial mediation on the relation between transformational leadership and intrinsic motivation. This indicated that some, but not all, of the effects of transformational leadership on intrinsic motivation occurred through trust and value congruence. 

Summary of Findings 

Table 13 summarises the findings of the study. 


Table 13: Summary of Multivariate Results 

 
SGF-Table13.png
 

Transactional Leadership Predicting Volunteer Extrinsic Motivation 

The results of hierarchical multivariate regression analysis indicated that the transactional behaviours of non-profit leaders significantly predict volunteer extrinsic motivation, but that transformational leadership behaviours do not do so. Correlation analysis also indicated a significant positive correlation between transactional leadership and extrinsic motivation. These results are consistent with transactional leadership theory which posits that such leadership establishes an exchange relationship with followers offering rewards for services rendered (Bass, 1985; Burns, 1978). They are also consistent with self-determination theory which connects extrinsic motivation to external and instrumental motivators (Deci and Ryan, 2000, 2008). The practice of transactional behaviours by non-profit leaders is likely to reinforce extrinsic motivation among volunteers. 

It is noteworthy that survey subjects rated senior pastors low in transactional leadership (M = 1.69) and high in transformational leadership (M = 3.24), and themselves low in extrinsic motivation (M = 1.27) and high in intrinsic motivation (M = 3.80). These results indicate that volunteers perceived senior pastors as more typically exercising transformational rather than transactional leadership behaviours and themselves as more typically experiencing intrinsic rather than extrinsic motivation. 

Transactional Leadership, Trust, and Value Congruence Predicting Volunteer Extrinsic Motivation 

The results of hierarchical multivariate regression analysis indicate that the transactional behaviours of leaders did not significantly predict volunteer trust in or value congruence with leaders. Correlation analysis also indicated no significant positive correlation between transactional leadership and trust or value congruence. Because transactional leadership did not have a statistically significant effect on trust or value congruence, these two variables did not have a mediating effect on the relationship between transactional leadership and extrinsic motivation. 

These results are not consistent with the findings of Jung and Avolio (2000) that transactional leadership predicts trust and value congruence and that these variables mediate the impact of transactional leadership on follower performance. They are also not consistent with transactional leadership theory which posits that transactional leaders build follower trust and value congruence by engaging in consistent behaviour, honouring agreements, sharing mutual values, and acknowledging the abilities and contributions of followers (Bass, 1985; Podsakoff et al., 1990). 

The absence of a relationship between transactional leadership and trust and value congruence is likely to be related to the low mean obtained for transactional leadership (M = 1.69). Survey subjects perceived senior pastors as more typically engaging in transformational (M = 3.24) rather than transactional leadership behaviours. It is possible that this significant difference in ratings of transformational and transactional leadership is in part related to an implicit leadership theory commonly held by volunteers (Offermann, Kennedy, and Wirtz, 1994). Transactional behaviours are practiced by effective leaders in order to enable volunteers to clearly understand their task descriptions and the desired outcomes that result from the performance of these tasks (Butler and Herman, 1999; Druskat, 1994; Son, 2003). This is an integral component of organisational leadership which is augmented by transformational behaviours (Bass, 1985; Judge and Piccolo, 2004; Rowold and Rohmann, 2009) and should likely have been rated more highly by subjects (Onnen, 1987). 

Offermann et al. (1994) observed that “individuals possess their own naïve, implicit theories of leadership and are readily willing to determine their boundaries and characteristics” (p. 44). Such theories are developed through socialisation and past experiences, and represent cognitive schemas specifying traits and behaviours that followers expect from leaders. They are stored in memory and activated when followers interact with a person in a leadership position (Epitropaki and Martin, 2004). These implicit leadership theories provide the basis for follower understanding of and response to leader behaviour. The possibility that transformational leadership is the implicit leadership theory for senior pastors among congregational members is consistent with previous findings that church volunteers prefer clergy to exercise transformational leadership behaviours (Bae, 2001; Druskat, 1994; Callahan, 1996; Choi, 2006; Onnen, 1987; Rowold, 2008) and with Christian understandings of leadership as being charismatic and virtuous (e.g., 1 Corinthians 11:1; Barrett, 1971; Fee, 1987; Morris, 1980). 

The existence of an implicit theory of transformational leadership held by church volunteers in regard to clergy is supported by Onnen’s (1987) finding that laity in churches seemed to perceive all clergy as being transformational even when the church was not growing or was declining. She suggested that a “halo” effect was likely to be present when laity rated clergy. A halo effect is present when a rater has a general conception of a person as being at a certain level and therefore possessing certain qualities. In the case of clergy, the role is identified with charisma and virtue as ministers seek to motivate laity through weekly preaching and related activities and are understood to exemplify Christian values. The position of senior pastor is likely to attract a “halo” which results in higher transformational and lower transactional ratings of the incumbent by volunteers, regardless of the actual performance of the senior pastor. 

Transformational Leadership Predicting Volunteer Intrinsic Motivation 

The results of hierarchical multivariate regression analysis indicated that the transformational behaviours of non-profit leaders significantly predict volunteer intrinsic motivation, but that transactional leadership behaviours do not do so. Correlation analysis also indicated a significant positive correlation between transformational leadership and intrinsic motivation. These results are consistent with transformational leadership theory which posits that such leadership establishes a moral and inspirational relationship with followers which motivates them to work for transcendental goals and for aroused higher-level needs for self-actualisation (Bass, 1985; Burns, 1978). They are also consistent with self-determination theory which connects intrinsic motivation to internal motivators to fulfil the needs for autonomy, competence, and relatedness (Deci and Ryan, 2000, 2008). The practice of transformational behaviours by non-profit leaders is likely to reinforce intrinsic motivation among volunteers. 

Transformational Leadership, Trust, and Value Congruence Predicting Volunteer Intrinsic Motivation 

The results of regression analysis indicate that the transformational behaviours of non-profit leaders significantly predicted volunteer trust in and value congruence with leaders. Correlation analysis also indicated significant positive correlations between transformational leadership and trust and value congruence. Further regression analysis indicated that both trust and value congruence exerted a partial mediation on the relation between transformational leadership and intrinsic motivation. This indicated that some, but not all, of the effects of transformational leadership on intrinsic motivation occurred through trust and/or value congruence. 

These results are consistent with the findings of Podsakoff et al. (1990), Shamir et al. (1993), and Jung and Avolio (2000) that transformational leadership predicts trust and value congruence and that these variables mediate the impact of transformational leadership on follower performance. They are also consistent with transformational leadership theory which posits that transformational leaders inspire follower trust and value congruence by means of personal example and vision-casting (Bass, 1985; Bennis and Nanus, 1985; Podsakoff et al. 1990). 

Survey subjects rated leaders high in transformational leadership (M = 3.24), and themselves high in intrinsic motivation (M = 3.80), trust (M = 3.63), and value congruence (M = 3.62). As well as the significant positive correlations between transformational leadership and trust and value congruence, there were also significant positive correlations between trust and value congruence, trust and intrinsic motivation, and value congruence and intrinsic motivation. These results indicate that volunteers perceived leaders as typically exercising transformational leadership behaviours and themselves as typically experiencing trust in and value congruence with leaders, as well as intrinsic motivation. 

Overview of Findings 

Transactional leadership has a positive relationship with volunteer extrinsic motivation and transformational leadership has a positive relationship with intrinsic motivation. Volunteer trust in and value congruence with leaders has a partial mediation effect on the relationship between transformational leadership and intrinsic motivation, but not on the relationship between transactional leadership and extrinsic motivation. 

These results are consistent with transactional and transformational leadership theory (Bass, 1985; Burns, 1978) and with self-determination theory (Deci and Ryan, 2000, 2008). They are also consistent with previous studies which have found that transactional and transformational leadership have positive organisational outcomes in non-profit organisations (Bae; 2001; Balswick and Wright, 1988; Butler and Herman, 1999; Callahan, 1996; Catano et al., 2001; Choi, 2006; Crain-Gully, 2003; Druskat, 1994; Johnson, 2007; Knudsen, 2006; Larsson and Ronnmark, 1996; Onnen, 1987; Rowold, 2008; Rowold and Rohmann, 2009; Son, 2003), and that trust and value congruence have mediating effects on the impact of transformational leadership on follower performance (Jung and Avolio, 2000; Podsakoff et al., 1990; Shamir et al., 1993). 

The contribution of this study has been to identify positive relationships between leadership behaviours and volunteer motivation, as well as mediating effects of trust and value congruence. 

Revised Study Model 

A revised study model is presented in Figure 2. It emphasises the relationship between the transactional and transformational leadership behaviours and the extrinsic and intrinsic motivation of volunteers. The relationship between transformational leadership and intrinsic motivation can be direct or mediated by volunteer trust in and/or value congruence with the leader. The relationship between transactional leadership and extrinsic motivation is direct and not mediated. 

 
SGF-Fig2.png
 

Figure 2: A revised model of the relationships between transactional and transformational leadership and volunteer motivation, with volunteer trust in and value congruence with the leader as mediating variables. 

Theoretical Implications 

This study contributed to the leadership and motivation literature by: (a) connecting transactional and transformational leadership theory with self-determination theory in order to examine the relationship between leadership behaviour and volunteer motivation in non-profit organisations, and thereby demonstrating (b) positive relationships between transactional leadership and extrinsic motivation and transformational leadership and intrinsic motivation, and (c)mediation effects of trust and value congruence on the relationship between transformational leadership and intrinsic motivation. 

No previous study had examined relationships between transactional and transformational leadership and volunteer motivation in a non-profit organisation. Self-determination theory had previously linked volunteer intrinsic motivation to situational variables including social environment and work context (Deci et al., 1994; Deci and Ryan, 2000, 2008). Environmental conditions that supported satisfaction of volunteers’ needs for autonomy, competence, and relatedness were found to predict intrinsic motivation. Millette and Gagné (2008) suggested that leadership behaviour was likely to be an important factor impacting on volunteer motivation while noting that such research had not been conducted. This study employed self-determination theory to conceptualise volunteer motivation, and demonstrated that transactional leadership predicts extrinsic motivation and transformational leadership predicts intrinsic motivation. This finding builds on and adds to existing understanding that transactional and transformational leadership have positive organisational outcomes in non-profit organisations (Bae; 2001; Balswick and Wright, 1988; Butler and Herman, 1999; Callahan, 1996; Catano et al., 2001; Choi, 2006; Crain-Gully, 2003; Druskat, 1994; Larsson and Ronnmark, 1996; Johnson, 2007; Knudsen, 2006; Onnen, 1987; Rowold, 2008; Rowold and Rohmann, 2009; Son, 2003). 

While previous studies had identified that trust and value congruence increase followers’ responsiveness to a leader and task performance (Jung and Avolio, 2000; Podsakoff et al., 1990; Shamir et al., 1993), none had investigated mediation effects of these variables on the relationship between leadership behaviour and volunteer motivation. This study demonstrated mediation effects of trust and value congruence on the positive relationship between transformational leadership and intrinsic motivation. The relationship between transformational leadership and intrinsic motivation and the mediation effects of trust and value congruence indicate that transformational leadership augments the impact of transactional leadership on volunteer motivation. Such an augmentation effect had previously been demonstrated in relation to follower performance (Bass, 1985; Judge and Piccolo, 2004), but not to volunteer motivation. The exercise of transformational leadership behaviours will augment the impact of transactional leadership behaviours on volunteer motivation in non-profit organisations by inspiring trust, value congruence, and intrinsic motivation. 

Practical Implications 

Non-profit organisations that depend on volunteer workers require leaders who can inspire intrinsic motivation in volunteers (Larsson and Ronnmark, 1996; Riggio et al., 2004). The organisational problems of shorter tenure and poorer task performance are less likely to occur among volunteers in non-profit organisations where leaders exercise transformational leadership behaviours directed towards the enhancement of volunteer trust, value congruence, and intrinsic motivation. Therefore, this study provides the following practical implications for non-profit organisations: (a) leader selection criteria should incorporate evidence of effective demonstration of transformational leadership behaviours; (b) leader training should incorporate transformational leadership behaviours that enhance volunteer trust, value congruence, and intrinsic motivation; and (c) leadership strategies should incorporate the goal of building volunteer intrinsic motivation. 

Self-determination theory posits that volunteers are motivated to satisfy their innate psychological needs for autonomy, competence, and relatedness. As these needs are met through voluntary activity, a volunteer is more likely to be intrinsically motivated and to provide higher quality task performance over a longer period of time (Deci and Ryan, 2008; Millette and Gagné, 2008). This study has shown that transformational leadership predicts trust, value congruence, and intrinsic motivation, and augments the impact of transactional leadership. Therefore, evidence of effective demonstration of the four interrelated transformational leadership behaviours of idealised influence, inspirational motivation, intellectual stimulation, and individualised consideration should be added to the transactional leadership behaviours of contingent reward and management by exception (Yammarino and Bass, 1990) as core criteria for leader selection. 

Training in transformational leadership behaviours should be provided to intending and incumbent organisational leaders. The findings of this study provide components of a transformational leadership training program. It should include training in how transformational leaders motivate volunteers and enhance their task performance. This would counter the tendency to mystify transformational leadership by mainly focusing on the leader’s charismatic qualities and personal characteristics (Jung and Avolio, 2000). Leaders should also be educated in self-determination theory and assisted in devising and implementing strategies to inspire volunteers’ intrinsic motivation. Practical training in critical behaviours that build trust, enhance value congruence, and increase intrinsic motivation should be built into a transformational leadership training program. 

Study Strengths and Weaknesses 

A key strength of this study is that it adds to the field of research regarding leadership behaviour and volunteer motivation. Previous research had not studied transactional and transformational leadership behaviours as predictors of volunteer motivation (Millette and Gagné, 2008). This study was able to identify positive relationships between leadership behaviours and extrinsic and intrinsic motivation among volunteers. 

Another strength of this study was the sample which consisted of 790 subjects from 28 different churches across a range of congregational sizes and rating the leadership behaviours of 28 different senior pastors. Although the congregations were selected to represent different size categories, the subjects were self-selected. The sample size and its representativeness suggest the findings may generalise to the broader population of non-profit organisations. 

A general weakness of this study was the self-reporting of volunteers’ motivation. The low rating of extrinsic motivation and the high rating of intrinsic motivation across subjects might indicate that the existence of socially desirable responses. In order to limit social desirability response bias, subjects were guaranteed response confidentiality and anonymity. However, some subjects may have responded according to who they would like to be rather than who they actually are. If so, this would likely result in higher intrinsic motivation self-ratings and lower extrinsic motivation self-ratings. 

A related weakness was the possible existence of an implicit theory of transformational leadership—a “halo” effect—at work in volunteers as they rated the leadership of senior pastors. It is likely that subjects gave higher ratings of senior pastors’ transformational leadership and lower ratings of transactional leadership because of their implicit perceptions of the appropriate characteristics of the position of senior pastor. 

Another weakness was the quantitative and cross-sectional design of the survey. As a quantitative study, it lacked the qualitative data that might be obtained from interviews and observations. This placed a limit on understanding why subjects provided the ratings that they did. The cross-sectional design did not allow for changes in the key variables and relationships over time. 

Directions for Future Research 

The results of this study indicate that future research should examine the existence of implicit theories of leadership held by congregational members in regard to clergy and the impact of such perceptions on their ratings of transactional and transformational leadership behaviours. Studies incorporating leaders’ self-ratings and peer ratings, as well as volunteer ratings, of non-profit organisation leaders’ behaviours are likely to provide greater insight into the exercise of transactional and transformational leadership behaviours. 

The relationships between transactional and transformational leadership, trust and value congruence, and extrinsic and intrinsic motivation could be further explored by conducting similar studies in other non-profit and for-profit organisations. Such studies across other organisations and industries would be needed to verify the findings of this study and to demonstrate whether they can be generalised. 

A longitudinal study could examine whether the relationships change over a longer period of time with increasing incumbent tenure. A related longitudinal study could examine whether interventions in the form of transformational leadership training produced stronger relationships with trust, value congruence, and intrinsic motivation. 

Finally, future studies should include qualitative research in a mixed methods design with interviews among volunteers in non-profit settings. In-depth interviews of a subsample of subjects would provide a richer perspective on why volunteers continue in their roles and what leadership behaviours contribute to the enhancement of intrinsic motivation. 

References 

Australian Christian Churches (2010). ACC 2010 church census report. Sydney, Australia: ACC National Office. 

Avolio BJ and Bass BM (1988) Transformational leadership, charisma and beyond. In Hunt JG, Baliga BR, Dachler HP and Schrieisheim CA (Eds.), Emerging leadership vistas. Emsford, NY: Pergamon Press, pp.29-50. 

Avolio BJ and Bass BM (1995) Individual consideration viewed at multiple levels of analysis: A multi-level framework for examining the diffusion of transformational leadership. Leadership Quarterly, 6: 199-218. 

Avolio BJ and Bass BM (2004) Multifactor leadership questionnaire: Manual and sampler set (3rd ed.). Palo Alto, CA: Mind Garden. 

Avolio BJ, Bass BM and Jung DI (1999) Re-examining the components of transformational and transactional leadership using the Multifactor Leadership Questionnaire. Journal of Occupational and Organizational Psychology, 72: 441-462. 

Bae K (2001) Transformational leadership and its application in church organization (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3005838). 

Balswick J and Wright W (1988) A complementary-empowering model of ministerial leadership. Pastoral Psychology, 37(1): 1-14. 

Barrett CK (1971) The first epistle to the Corinthians (2nd ed.). London, UK: A & C Black. 

Bass BM (1985) Leadership and performance beyond expectations. New York, NY: Free Press. 

Bass BM (1998) Transformational leadership: Industry, military, and educational impact. Mahwah, NJ: Erlbaum. 

Bass BM and Avolio B J (1990) The implications of transformational and transactional leadership for individual, team, and organizational development. In Woodman R and Passmore W (Eds.) Research in organizational change and development (pp. 231-272). Greenwich, CT: JAI Press. 

Bass BM and Avolio BJ (2000) Effects on platoon readiness of transformational/transactional platoon leadership. Ft Belvoir, VA: US Army Research Institute for the Behavioral and Social Sciences. 

Bennis W and Nanus B (1985) Leaders. New York, NY: Harper & Row. 

Burns JM (1978) Leadership. New York, NY: Harper & Row. 

Butler DM and Herman RD (1999) Effective ministerial leadership. Nonprofit Management & Leadership, 9(3): 229-239. 

Callahan SH (1996) A delphi study of the competencies needed by leaders of Roman Catholic faith communities in western Washington through the year 2000 (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 9716963). 

Catano VM, Pond M and Kelloway EK (2001) Exploring commitment and leadership in voluntary organizations. Leadership and Organization Development Journal, 22(5/6): 256-263. 

Choi K (2006) The impact of motivational language and transformational leadership of the pastor on volunteers’ job satisfaction in Korean Southern Baptist churches in the United States (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3243509). 

Crain-Gully DI (2003) Exploring transformational leadership characteristics among African American church leaders in the San Francisco bay area (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3082000). 

Deci EL, Eghrari H, Patrick BC, and Leone DR (1994) Facilitating internalization: The self determination theory perspective. Journal of Personality, 61(1): 119-142. 

Deci EL and Ryan RM (2000) The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11(4): 227-268. 

Deci EL and Ryan RM (2008) Self-determination theory: A macrotheory of human motivation, development, and health. Canadian Psychology, 49(3): 182-185. 

Druskat VU (1994) Gender and leadership style: Transformational and transactional leadership in the Roman Catholic Church. Leadership Quarterly, 5: 99-119. 

Epitropaki O and Martin R (2004) Implicit leadership theories in applied settings: Factor structure, generalizability, and stability over time. Journal of Applied Psychology, 89(2): 293-310. 

Fee GD (1987) The first epistle to the Corinthians. Grand Rapids, MI: Eerdmans. 

Hair JF, Black WC, Babin BJ and Anderson RE (2010) Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall. 

Johnson B (2007) The perceived leadership crisis within the Baptist Church: An exploratory empirical investigation of selected churches in Memphis, Tennessee (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3251346). 

Judge TA and Bono JE (2000) Five-factor model of personality and transformational leadership. Journal of Applied Psychology, 85(5): 751-765. 

Judge TA and Piccolo RF (2004) Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89, 755-768. 

Jung DI and Avolio BJ (2000) Opening the black box: An experimental investigation of the mediating effects of trust and value congruence on transformational and transactional leadership. Journal of Organizational Behavior, 21(8): 949-964. 

Knudsen DN (2006) The effect of transformational leadership on growth in specialized non-profit organizations: Churches (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3231161). 

Larsson S and Ronnmark L (1996) The concept of charismatic leadership: Its application to an analysis of social movements and a voluntary organization in Sweden. The International Journal of Public Sector Management, 9(7): 32-41. 

Lichtman SL and Malony HN (1990). Effective ministerial style as perceived by denominational leadership. Pastoral Psychology, 38(3): 161-171. 

Millette V and Gagné M (2008) Designing volunteers’ tasks to maximize motivation, satisfaction and performance: The impact of job characteristics on volunteer engagement. Motiv Emot, 32: 11-22. 

Morris L (1980) The first epistle of Paul to the Corinthians. Grand Rapids, MI: Eerdmans. 

Offermann LR, Kennedy JK and Wirtz PW (1994) Implicit leadership theories: Content, structure, and generalizability. Leadership Quarterly, 5(1), 43-58. 

Onnen MK (1987) The relationship of clergy leadership characteristics to growing or declining churches (Unpublished doctoral dissertation). University of Louisville, Louisville, KY. 

Pallant J (2011) SPSS survival manual (4th ed.). Crows Nest, Australia: Allen & Unwin. 

Paternoster R, Brame R, Mazerolle P and Piquero A (1998) Using the correct statistical test for the equality of regression coefficients. Criminology, 36(4): 859-866. 

Phillips LC and Phillips MH (2010) Volunteer motivation and reward preference: An empirical study of volunteerism in a large, not-for-profit organization. SAM Advanced Management Journal, 75(4): 12-19. 

Phillips L and Phillips M (2011, Winter) Altruism, egoism, or something else: Rewarding volunteers effectively and affordably. Southern Business Review: 23-35. 

Podsakoff P, Mackenzie S, Moorman R and Fetter R (1990) Transformational leader behaviors and their effects on followers’ trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly, 1: 107-142. 

Posner B (1992) Person–organization values congruence: No support for individual differences as a moderating influence. Human Relations, 45(4): 351-362. 

Posner BZ (2010) Another look at the impact of personal and organizational values congruency. Journal of Business Ethics, 97: 535-541. 

Posner BZ, Kouzes JM and Schmidt WH (1985) Shared values make a difference: An empirical test of corporate culture. Human Resource Management, 24(3): 293-310. 

Riggio RE, Bass BM and Orr SS (2004) Transformational leadership in nonprofit organizations. In: Riggio RE and Orr SS (eds.), Improving leadership in nonprofit organizations. San Francisco, CA: Jossey-Bass, pp.49-62. 

Rowold J (2008) Effects of transactional and transformational leadership in pastors. Pastoral Psychology, 56: 403-411. 

Rowold J and Rohmann A (2009) Relationships between leadership styles and followers’ emotional experience and effectiveness in the voluntary sector. Nonprofit and Voluntary Sector Quarterly, 38(2): 270-286. 

Shamir B, House R and Arthur M (1993) The motivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4(2): 1-17. 

Snyder M and Omoto AM (2004) Volunteers and volunteer organizations: Theoretical perspectives and practical concerns. In Riggio RE and Orr SS (Eds.), Improving leadership in nonprofit organizations (pp. 163-179). San Francisco, CA: Jossey-Bass. 

Son J (2003) An analysis of leadership styles and practices among Korean senior pastors (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3080173). 

Wilson J and Janoski T (1995) The contribution of religion to volunteer work. Sociology of Religion, 56(2): 137-152. 

Wilson J and Musick M (1997) Who cares? Toward an integrated theory of volunteer work. American Sociological Review, 62(5): 694-713. 

Yammarino FJ and Bass BM (1990) Transformational leadership and multiple levels of analysis. Human Relations, 43(10): 975-995. 

Yammarino FJ, Spangler WD and Bass BM (1993) Transformational leadership and performance: A longitudinal investigation. Leadership Quarterly, 4: 81-102. 

Yukl G (2006) Leadership in organizations (6th ed.). Upper Saddle River, NJ: Pearson Prentice Hall. 

Bel Litchfield