Transformational Leadership,Goal Setting,and Work Motivation

Abstract

Many studies have recently been devoted to transformational leadership in public sector organizations. However, these seldom address the relationship between transformational leadership and work motivation. By combining separate findings in the public administration literature about transformational leadership style and goal setting and about the work motivation of public employees, we construct hypotheses on how transformational leadership style and work motivation are related in a public sector context. Building on this, our analysis, based on a survey in a Dutch municipality, shows several linkages between this leadership style and work motivation. In addition to a direct effect, we anticipated and indeed found two indirect ones. First, goal setting mediates this relationship, with transformational leaders able to set more challenging and more specific goals. Second, transformational leaders seem able to reduce perceptions of procedural constraints among their employees. This is especially important as such constraints can hamper the use of goal setting.

Keywords

goal setting public sector work context transformational leadership work motivation

Introduction

Motivation is an important determinant of job performance, and hence it is a key element when studying fields such as HRM, Organizational Behavior, and Organizational Psychology (Latham, 2007). Motivation is also an important topic within public administration, but with respect to the motivation of public employees, Rainey (2009, p. 243) suggests that it is “greatly affected by the public sector environment”.

In a meta-review, Perry, Mesch, and Paarlberg (2006) discuss four antecedents of public employees’ work motivation: employee incentives, job design, goal setting, and employee participation. Although all these antecedents are important, Perry et al. (2006, p. 509) note that goal setting is “the single most researched and dominant theory of employee motivation in the field.” At the same time, they observe that the application of this theory in public settings is limited.

In this respect, the work of Wright (2001, 2004) is an important exception. Based on his findings, Wright (2004, p. 73) argues that within a public sector context, it is also important to set goals that are “specific, difficult, and doable.” However, he also notes that certain work characteristics (e.g., organizational goal conflicts, procedural constraints, and goal ambiguity) negatively affect antecedents of work motivation. Although such characteristics are not unique to public sector organizations, these organizations probably have to deal with them more often (Rainey, 2009, p. 83-85). Wright’s studies thus illustrate the effect of the public sector environment on motivation.

The question arises as to what this means for public sector managers and their leadership style. Whereas Wright (2004) uses goal setting variables as independent antecedents of work motivation, several other studies have considered the influence of leadership style on elements of goal setting within a public context. For instance, Moynihan, Pandey, and Wright (2012) showed that transformational leaders are able to influence goal clarity. Similarly, Wright, Moynihan, and Pandey (2012) showed that transformational leadership boosts employee mission valence by fostering employees’ public service motivation and perceptions of goal clarity.

It is no coincidence these recent studies have focused on the transformational leadership style. In the last decade, many public administration studies have addressed this style (Van Wart, 2013). Although studying other styles—for instance forms of distributive leadership (Van Wart, 2012)—is certainly relevant, we will limit ourselves here also to the transformational style as the motivation of individuals—our main object of the study—is an important aim of transformational leadership to achieve higher levels of performance (Bass, 1985). With respect to our topic, it has been suggested that goal setting relates this leadership style with motivation (Ilies, Judge, & Wagner, 2006).

In this article, we therefore combine separate public sector findings on (a) goal setting and motivation and (b) transformational leadership and the setting of goals. Such a combination could provide an answer to the question as to how the goal setting process can be influenced in a public sector context. Against this background, we set out to answer the following research question: How does a transformational leadership style affect the work motivation of employees working in a public sector context?

The general literature on transformational leadership in the private sector has shown a positive effect of this type of leadership on work motivation (Dvir, Eden, Avolio, & Shamir, 2002; Masi & Cooke, 2000). We will argue in this article that transformational leadership affects the work motivation of public sector employees in several ways. Reducing organizational goal conflict and red tape is one. This reduction will positively influence employees’ goal-setting processes, which in turn will positively affect their work motivation.

To answer our research question, the existing theoretical framework is further developed in the second section of the article, resulting in four hypotheses. In the third section, our data and measurements are discussed. Next, in the fourth section, our results—based on structural equation modeling (SEM)—are presented. We show that the results support our initial model (albeit with some modifications). The article ends by drawing conclusions and discussing some of the limitations of the research.

Theoretical Framework

This section is structured as follows. First, we will explain why we expect a direct relationship between transformational leadership and work motivation. Following this, we will hypothesize that two mechanisms relate these two concepts. In the first place, transformational leadership will positively affect the goal-setting process, which in turn will affect work motivation. In the second place, transformational leadership will indirectly affect the setting of goals in a public sector work context by reducing red tape and organizational goal conflict.

Transformational Leadership and Work Motivation

The focus of this article is on work motivation, which has choice, effort, and persistence at the heart of its definition (Latham, 2007). Although Perry et al.’s (2006) meta-review did not identify leadership style as an important antecedent of work motivation, many others have stated that motivation is related to leadership behavior or, in the words of Latham (2007, p. 4), “motivation is a core competence of leadership.”

Although there are many theories concerning leadership (Van Wart, 2012), transformational leadership has become an important focus of interest (Burns, 1978; Dvir et al., 2002; Van Wart, 2012). The basic idea behind this leadership concept is that “effective leaders transform or change the basic values, beliefs, and attitudes of followers so that they are willing to perform beyond the minimum levels specified by the organization” (Podsakoff, MacKenzie, Moorman, & Fetter, 1990, p. 108). This emphasizes a leadership style that is inspirational (Van Wart, 2012) and is “centered on the assumption that leaders can change followers’ beliefs, assumptions, and behavior by appealing to the importance of collective or organizational outcomes” (Moynihan, Pandey, & Wright, 2012, p. 147).

Many studies on transformational leadership have stressed its positive effects, for instance, on performance (Wang, Oh, Courtright, & Colbert, 2011) and employee outcomes (Whittington, Goodwin, & Murray, 2004). Although we are also looking here at the positive effects, it should be noted that several studies have stressed negative effects. For instance, Harrison (1987) warned of a burn-out risk and Stephens, D’Intino, and Victor (1995) noted that the theory is biased toward certain stakeholders (top management and customers) at the expense of employees. Other critics have warned of the over-emphasis on universal applicability (van Wart, 2012).

Yukl (1999) noted that within transformational leadership theory, it is often not clear which mechanisms explain how such leaders influence followers’ behavior. Nevertheless, a positive effect on work motivation can be expected based on the elements that, according to the theory, are central to transformational leadership (van Wart, 2012). A transformational leader is expected (a) to show individualized consideration by diagnosing and elevating the needs of each follower, (b) to become a source of admiration (idealized influence), (c) to stimulate their followers to see the world from new perspectives (intellectual stimulation), and (d) to provide inspirational motivation and thus meaning and a sense of purpose about what needs to be done. Given these qualities, a transformational leadership style will positively affect work motivation by enhancing the choices made by employees in terms of devoting effort to certain tasks and their willingness to persist in these. Several studies have already shown that transformational leadership is indeed related to work motivation (Masi & Cooke, 2000; Shamir, House, & Arthur, 1993). As we are interested in the mediating effects of goal setting, we nevertheless need to establish whether there is also in our (public sector) study a direct effect of transformational leadership on work motivation. Our first hypothesis is therefore

Hypothesis 1: A transformational leadership style is positively related to the work motivation of public sector employees.

Goal Setting as a Mediator Between Transformational Leadership and Work Motivation

Locke and Latham (2002) developed the high performance cycle to describe the relationship between goal setting and performance. However, not all goals will influence performance, and their potential effect is moderated by goal commitment, their self-efficacy, the degree to which feedback is provided, and the complexity of the tasks to be performed (Locke & Latham, 2002). In respect of the last of these influences, it should be noted that tasks should be challenging but not too difficult: people have to believe that their objectives are achievable. Also, goals should be specific not vague. Thus, goal difficulty and goal specificity are important aspects of goal setting.

How is this related to leadership style? We will argue that transformational leadership affects goal specificity and goal difficulty. As goal setting affects work motivation, goal setting will thus act as a mediator between transformational leadership and work motivation.

Only a few studies have addressed how transformational leadership, goal setting, and motivation are related. Here, Kirkpatrick and Locke (1996) suggested that setting goals could be an explanatory mechanism between transformational leadership and followers’ performance. One explanation they offer is that goals (and also self-efficacy and performance) form a “motivation hub.” Goals intervene between the vision set by a leader and the action taken by a follower. This accord with one of the two mechanisms intervening between transformational leadership and work motivation described by Ilies et al. (2006) who argue that emotional and cognitive processes link the two concepts. The setting of goals is part of the cognitive process. The visions presented by transformational leaders form the linking pin: such a vision can be seen as an overarching idealized goal, which through a cascade process is translated into more specific and concrete goals.

Other studies describe the linking mechanism between transformational leadership and goal setting in more detail. For instance, Berson and Avolio (2004) describe how transformational leaders will use individualized consideration (one of the elements of transformational leadership mentioned above) to understand how followers understand messages sent from above and then “adjust these messages to the level of the follower’s understanding.” In this respect they show that transformational leaders are effective communicators. Whittington et al. (2004, p. 596) similarly argue that leaders “translate their visions into tangible contributions ( . . . ) through the goal setting process.” In other words, transformational leaders make organizational goals clear to employees by specifying them.

Podsakoff et al. (1990) describe how transformational leadership and goal difficulty are linked. Based on their literature review, they conclude that one of the key elements of transformational leadership behavior is “setting high performance expectations.” High expectations imply more difficult goals. Whittington et al. (2004) use a similar argument in stating that transformational leaders use challenging goals to communicate high expectations.

Transformational leaders will thus set more specific and more difficult goals than other leaders. As goal setting is, in turn, related to work motivation, our hypothesis is

Hypothesis 2: Goal setting positively mediates the relationship between a transformational leadership style and the work motivation of public sector employees.

Goal Setting and Work Motivation in a Public Context

We now turn to aspects of the public sector as a work context that could affect goal setting. As far back as 1982, Perry and Porter suggested that goal-setting theory could be relevant to the public sector, but cautioned concerning the often vague and conflicting nature of goals in such organizations. Latham, Borgogni, and Petitta (2008) concurred and stated that vague goals could indeed reduce the effectiveness of goal setting in a public sector context. These observations fit within the debate as to what makes public sector organizations distinctive. In this respect, it is often argued that there is a greater vagueness in, and multiplicity of, organizational goals, and also greater organizational goal conflict, within public organizations when compared with private ones (Rainey, 2009). It is also argued that public organizations experience red tape, which can hamper their performance (Brewer & Walker, 2010; Walker & Brewer, 2009).

There is a debate as to what extent public and private organizations really differ (Rainey, 2009, pp. 58-88). Indeed, employees of private organizations could also experience vague goals, organizational goal conflict, and red tape (Rainey, Pandey, & Bozeman, 1995). For our research, the extent of the differences between public and private organizations as such is not important, it is sufficient that these conditions shape the work environment of many public employees, albeit probably to varying degrees. In such circumstances, using goal setting as an instrument to positively influence work motivation is not straightforward. In the Introduction, we referred to Wright (2001, 2004) who tested a conceptual framework on how elements of the public sector work context affect work motivation. Based on this, we will test the following hypothesis:

Hypothesis 3: The work context of public sector organizations has a negative indirect effect on the work motivation of public sector employees through its negative effect on goal setting.

Transformational Leadership and Goal Setting in a Public Sector Context

Wright (2004), in his study on goal setting and work motivation in the public sector, did not consider leadership characteristics. However, other studies suggest including these. In this respect, Walker and Brewer (2009) showed that public managers are able to reduce red tape by employing appropriate management techniques and strategies. Moynihan, Wright, and Pandey (2012) found that a transformational leadership style reduces the perception of human resources’ red tape whereas, in an earlier study, Wright and Pandey (2010) did not find such an effect. Several studies have also shown that transformational leaders are able to clarify goals. In this respect, Moynihan, Wright, and Pandey (2012), and Wright et al. (2012) showed that a transformational leadership style had a positive effect on goal clarity. Densten (2005) argued that the inspirational motivation delivered by transformational leaders leads to a reduction in goal conflict by establishing a collective sense of identity among followers.

The above discussion suggests an indirect relationship between a transformational leadership style and goal setting within public organizations. As transformational leaders will reduce the perception of conflicting organizational goals (such as by making clear which goals are most important) and/or red tape (for instance because followers will see constraints in a different light and thus be able to overcome them), it becomes easier to set challenging and specific goals. As such, a transformational leadership style will be indirectly related to goal setting in public organizations as this style will reduce elements that, within this context, negatively affect the goal-setting process. This leads to our fourth hypothesis:

Hypothesis 4: A transformational leadership style will have a positive indirect effect on goal setting through its effect on the public sector work context.

We summarize our expectations, posited in the four hypotheses, in the model shown in Figure 1.

Figure 1.

Conceptual model.

The model suggests how a transformational leadership style and the work motivation of public sector employees might be related. In the first place, we expect a direct relationship between this leadership style and work motivation because a transformational leadership style will directly inspire people. Second, there will be an indirect effect on work motivation, because transformational leaders are able to positively influence the goal-setting process by setting specific and challenging goals. This is important in a public sector context (our third point) because some elements that are often to be found there (such as organizational goal conflicts and red tape) have a negative effect on this goal-setting process, and thus on motivation. Furthermore, a transformational leadership style will reduce employees’ perceptions of organizational goal conflict and red tape, and this will positively influence the goal-setting process.

Methodology

Data

We conducted our research within the second largest municipality in the Netherlands (the city of Rotterdam with about 13,000 municipal employees). To minimize the possible effect of mixed work contexts, we limited our sample to “knowledge workers,” thereby excluding workers in, for example, refuse collection. The advantage gained (a fairly unified work context) has an obvious disadvantage in that it limits the possibility of generalizing the findings. As such, our findings will be limited to a specific work context within a part of the Dutch public sector work.

Following Davenport (2005), we define knowledge workers as those employees who have high levels of expertise, education, or experience, and whose jobs primarily involve the creation, distribution, or application of knowledge. This limited our population to about 6,500 employees within the “administrative support,” “policy,” “supervision,” and “project management” job categories. From these, a random sample of 3,105 workers was drawn. As some e-mail addresses were missing, we approached 3,049 respondents in April/May 2011. In this e-mail, they were asked to visit a webpage and answer an online questionnaire. In total, 1,069 employees responded. Unfortunately, not all fully completed the questionnaire, leaving 958 useable responses (an effective response rate of 31%).

Of these respondents, 44% were females, close to the female share of the employee population (45%). The mean age was 46.2 years, slightly above that in the organization as a whole (44.4 years). Distribution by job category and salary scale largely mirrored the population in the organization—although workers in lower job categories and on lower salary scales were somewhat underrepresented. In terms of job categories, 48% of the respondents held an administrative staff position, 17% were policy workers, 14% were working on a specific project, and 22% held supervisory positions.

Measurement

This section addresses the measurement of the variables included in the conceptual model. A full list of the items used is included in the appendix. Most of them are based on the early study by Wright (2004), which was used as a starting point for our own research. The use of a survey imposes the significant limitation that we can only measure the perceptions of employees. Furthermore, we do not compare public and private organizations. Consequently, in this article, we can only address the extent to which certain perceived public sector characteristics are important in the relationship linking transformational leadership, goal setting, and work motivation.

Work motivation

To measure work motivation, we used Wright’s six-item scale (2004). Respondents could respond to each item using a 5-point Likert scale ranging from totally disagree (1) to totally agree (5). Based on the results of a confirmatory factor analysis (CFA), two of the items (“I probably do not work as hard as others who do the same type of work” and “Time seems to drag while I am on the job”) were removed from this scale. All the standardized loadings were statistically significant, and the internal reliability of the scale was .67. Although slightly below the .70 threshold normally used for internal consistency, we considered this acceptable as the items are based on a scale that has been shown to be reliable in earlier research (Wright, 2004), and the fact that the value of Cronbach’s α, as the measurement of internal consistency, is also dependent on the number of items used (Field, 2005).

Goal setting

In our analysis, we limited the goal-setting process to two variables: job goal specificity and job goal difficulty. These were identified as important variables in Locke and Latham’s model (2007), but this means that we have not included potentially important moderators (such as feedback, task complexity, self-efficacy, and goal commitment). The reason for this was partly practical as we were concerned that the questionnaire would become overlong, reducing the response rate. Furthermore, including them would have also made our model much more complex and distracted from the main purpose of the research.

Job goal specificity

Four items were used to measure job goal specificity (Wright, 2004). Again, answers were to be given using a 5-point Likert scale ranging from totally disagree (1) to totally agree (5). Based on the results of the CFA, one of the items (“I understand fully which of my job duties are more important than others”) was removed from the scale. All the standardized loadings were statistically significant, and Cronbach’s α for the goal specificity measure was .69.

Job goal difficulty

The five items used to measure job goal difficulty were also based on Wright (2004), with answers given using the same 5-point Likert scale. All the standardized loadings were greater than .50, and all statistically significant. The internal reliability was .77.

Public sector work context

We included two variables also used by Wright (2004) to measure the public sector work context: organizational goal conflict and procedural constraints. The latter can be seen as closely related to the concept of red tape. As noted in the theoretical section above, these elements are likely to affect goal setting and motivation, and indeed were found to be important in Wright’s study (2004), although only with respect to job specificity.

Procedural constraints

This concept was measured using Wright’s (2004) five-item scale with the same 5-point Likert scale as above. All the standardized loadings were greater than .50, and they were all statistically significant. The internal reliability was .75.

Organizational goal conflict

Here, we used Wright’s (2004) four-item scale. Based on the results of the CFA, one of the items (“To satisfy some people, this organization will inevitably upset others”) was removed. The resulting three items all showed statistically significant standardized loadings greater than .50, and Cronbach’s α for the scale items was .81.

Transformational leadership

To measure transformational leadership, we used Carless, Wearing and Mann’s (2000) seven-item scale. Again, responses were matched to a 5-point Likert scale ranging from totally disagree (1) to totally agree (5). All the standardized loadings were greater than .50, and were statistically significant. Cronbach’s α for the transformational leadership measure was .94.

Control variables

Four control variables were included in the analyses: gender (1 = female), age, educational level, and supervisory position (1 = supervisory position). We included the latter two as Locke and Latham (1990) argued that the nature of the job could affect the goal-setting process. For instance, individuals with more demanding jobs are expected to exert greater effort and persistence. It seems reasonable to assume that the more highly educated employees and those in supervisory positions are more likely to have a demanding job.

Results

A two-step SEM approach was adopted for analyzing the data (Anderson & Gerbing, 1988). Here, the factor structure of all the variables in the study was examined, and the resulting model’s fit tested with a CFA. Following this, the proposed structural model was examined to assess the validity of the hypothesized relationships between the constructs. As three of our four hypotheses include mediation effects, we employed a bootstrapping method (Hayes, 2009). In its simplest form, this method estimates the parameters of a model and their standard errors from the sample alone, without reference to any theoretical sampling distribution. In our study, we created 5,000 samples (with replacement) from the available observed sample. From these samples, one can establish robust estimates of the expected value and the variability of the statistics (Hox, 2003). The CFA and the SEM were conducted using AMOS version 16. The overall fit of the models was evaluated using a combination of absolute and relative fit indices: a relative chi square (CMIN/DF), a goodness of fit index (GFI), a comparative fit index (CFI) and a root mean square error of approximation (RMSEA).

The Measurement Model

The first step in the Anderson and Gerbing (1988) two-step approach is to examine the overall factor structure of all the research variables. Based on the results of this CFA, the measurement model is then modified. Here, the modifications made to enhance the model included the introduction of some error correlations. The resulting measurement model provided an adequate fit to the data (CMIN/DF = 2.90, GFI = .92, CFI = .93, RMSEA = .047), with each indicator significantly loaded onto the appropriate factor and all loadings above .47. These results verify the posited relationships among the indicators and constructs, thus demonstrating the convergent validity of the constructs. Items related to the same construct were always more closely correlated with one another than with items from other constructs, and this is an indicator of discriminant validity. Furthermore, Bagozzi and Philips (1982) argue that discriminant validity is achieved in SEM if the unconstrained model has a significantly lower chi-square value than the constrained model. In this study, the chi-square value for the unconstrained model (CMIN = 1171.927/DF = 404) is significantly lower than for the constrained model (CMIN = 3963.592/DF = 425). Thus, for this model, discriminant validity has been demonstrated.

In Table 1, the means, standard deviations, and correlations between the constructs are presented. Almost all the correlations are significant at the .01 level. Only the correlation between organizational goal conflict and job goal difficulty is significant at the .05 level. Work motivation is correlated with all the other variables, and has an especially high correlation with job goal difficulty (.65). The correlation between transformational leadership and work motivation is also substantial (.33). The procedural constraints construct is negatively related to all the variables with the exception of organizational goal conflict, with our data suggesting that the higher the organizational goal conflict, the more procedural constraints there are. As some of the bivariate correlations are in the medium to high range, we conducted multicollinearity tests. The variance inflation factor (VIF) values were all well within the acceptable range (Bowerman & O’Connell, 1990), with the highest being 1.5. We can, thus, conclude that our results are not adversely affected by multicollinearity.

Table 1.

Means, Standard Deviations, and Correlations (N = 863).

	M	SD	1	2	3	4	5
1. Transformational leadership	3.30	0.87	—
2. Procedural constraints	2.79	0.75	−.425**	—
3. Organizational goal conflict	3.33	0.67	−.409**	.640**	—
4. Job goal difficulty	3.74	0.59	.255**	−.350**	−.100*	—
5. Job goal specificity	3.61	0.67	.441**	−.371**	−.407**	.226**	—
6. Work motivation	4.16	0.52	.333**	−.493**	−.272**	.652**	.345**

p < .05. **p < .01.

The Structural Model

To test the proposed relationships, a causal structure was posited that resulted in a SEM. The overall model was evaluated using several fit indices. The calculated model fit values were 2.95 (CMIN/DF), .93 (GFI), and .94 (CFI), all of which indicate that the model has an acceptable fit. Furthermore, the RMSEA value of .048 (PClose .870) is also indicative of a good model fit. In Figure 2 of the resulting model, only the significant relationships (with a statistical significance level of .05 or better) are shown. The numeric values indicated on the lines are the standardized regression coefficients (beta), and the values in parentheses are the explained variance.

Figure 2.

Results of structural equation modeling.

The next step in the analysis was to examine the control variables. Two of the control variables, age and educational level, were dropped from the final model as they did not have a significant influence. Thus, in the final model, the only control variables included were gender and supervisory position. This final model provided an adequate fit to the data (CMIN/DF = 3.03, GFI = .92, CFI = .93, RMSEA = .048 PClose = .769).

Model validity was checked through cross-model validation. Camilleri (2006) argues that this cross-validation process should include three phases. In the first phase, the data are divided into two data sets with a random selection of 20% and 80% respectively of the data collected. In the second phase, the structural fit index (R²) is calculated for both data sets. The third phase consists of examining the differences between the calculated structural fit indices for the two data sets. The extent of the model’s validity is determined by the similarity in the variances of each data set. The results of such a cross-model validation are presented in Table 2. The fact that the difference in the explained variances is small indicates that our model provides a valid description of the data.

Table 2.

Results of Cross-Model Validation Showing R² Values for the Three Samples.

	Full sample	80% sample	20% sample	Difference in R²
Organizational goal conflict	.169	.197	.147	.050
Red tape	.467	.449	.372	.077
Goal specificity	.265	.267	.283	−.016
Goal difficulty	.216	.237	.228	.009
Work motivation	.512	.526	.608	−.082

Looking at Figure 2, one would conclude that our results largely support the original model. However, as our hypotheses include mediation effects, we require a further test to see whether mediation does indeed take place.

Table 3 presents both direct and indirect effects. The direct effects are standardized regression coefficients similar to those shown in Figure 2. To test our mediation hypotheses, we used a bootstrapping procedure. In addition to robust estimates, bootstrapping procedures also provide bias-corrected confidence intervals that enable us to judge the significance of indirect effects. The indirect effects can be found in the lower half of the table. Furthermore, as we have used more than one variable in measuring our concepts (for instance, public sector context is measured with two variables: procedural constraints and organizational goal conflict), we are able to split the overall indirect effect into the specific effects of each variable. When testing our mediation hypotheses (2-4), we first consider the overall indirect effect before taking a closer look at the specific effects.

Table 3.

Direct and Indirect Effects.

Dependent variables
Independent variables	Work motivation	Goal specificity	Goal difficulty	Procedural constraints	Organizational goal conflict
Direct effects
Transformational leadership	.102* (.054) (H1)	.336** (.046)	.157** (.053)	−.176** (.046)	−.411** (.039)
Goal specificity	.184** (.062)	—	—	—	—
Goal difficulty	.623** (.044)	—	—	—	—
Procedural constraints	—	ns	−.383** (.082)	—	—
Organizational goal conflict	—	−.275** (.056)	.209** (.074)	.541** (.051)	—
Gender	.128** (.036)	ns	ns	.096* (.039)	ns
Supervisory position	ns	ns	.195** (.039)	−.184** (.040)	ns
Indirect effects
Transformational leadership (H2, H4)	Total indirect effect: .221** (.044) (H2)	Total indirect effect: .113** (.026) (H4)	Total indirect effect: .066** (.031) (H4)	—	—
Transformational leadership via goal setting (H2)	Total indirect effect via goal setting: .206** (.043) (H2)	—	—	—	—
	Indirect effect via goal difficulty: .133** (.034 ) (H2)
	Indirect effect via goal specificity: .099**(.033) (H2)
Transformational leadership via public sector work context (H4)	—	Indirect effect via procedural constraints: ns (H4)	indirect effect via procedural constraints: .076** (.027) (H4)	—	—
		Indirect effect via organizational goal conflict: .103** (.024) (H4)	Indirect effect via organizational goal conflict: ns (H4)
Procedural constraints via goal setting (H3)	Total indirect effect via goal setting: −.238** (.057) (H3)	—	—	—	—
	Indirect effect via goal difficulty: −.238** (.057) (H3)
	Indirect effect via goal specificity: ns (H3)
Organizational goal conflict via goal setting (H3)	Total indirect effect via goal setting: ns (H3)
	Indirect effect via goal difficulty: ns (H3)
	Indirect effect via goal specificity: −.070** (.028) (H3)

Note. ns = not significant.

p < .05. **p < .01.

First, we tested our hypothesis that a transformational leadership style is positively related to the work motivation of public sector employees. The results indeed show a positive direct relationship between these two variables (β = .102, SE = .054, p < .05), so the data support this hypothesis.

Second, we tested the hypothesis that goal setting mediates the relationship between a transformational leadership style and the work motivation of public sector employees. The results in the first row of Table 3 show a statistically significant indirect overall effect of a transformational leadership style on work motivation (β = .221, SE = .044, p < .01). Given that the relationship between transformational leadership and work motivation is mediated by two groups of intervening variables (through public sector context and goal setting), we cannot be sure which of these (or both) mediates this relationship. To clarify this, we tested for the indirect effect of transformational leadership through goal setting (without the intervening effect of public sector context) on work motivation. The results are presented in the second row of Table 3 and show that this indirect effect is statistically significant (β = .206, SE = .043, p < .01). Having determined that the indirect effect of goal setting is significant, we need to know whether the individual effects of the two goal-setting variables are both statistically significant. The results are presented in the second row of Table 3. They indicate that both goal specificity (β = .099, SE = .033, p < .01) and goal difficulty (β = .133, SE = .034, p < .01) mediate the relationship between transformational leadership and work motivation. Thus, the data support our second hypothesis.

Our third hypothesis argues that the work context of public sector organizations has a negative indirect effect on work motivation through its effect on goal setting. Table 3 shows that procedural constraints have a statistically significant overall indirect effect on work motivation (β = −.238, SE = .057, p < .01), supporting this hypothesis. However, having used two variables to measure goal setting and two to measure public sector work context, we have to test four specific indirect pathways. Even though the overall indirect effect of procedural constraints is statistically significant, the results show that the specific indirect effect of procedural constraints on goal specificity is not. In other words, only goal difficulty has a mediating effect (β = −.238, SE = .057, p < .01). Also, the overall indirect effect of organizational goal conflict on work motivation is not statistically significant. Overall, our results indicate that only one of the four possible mediating relationships between public sector context and work motivation is statistically significant: goal difficulty mediates the relationship between procedural constraints and work motivation. As such, the data only partly support Hypothesis 3.

Finally, we hypothesized that a transformational leadership style would positively affect the goal-setting process through its effect on the public sector work context. The results in the first row of Table 3 show a statistically significant indirect overall effect of transformational leadership on goal specificity (β = .113, SE = .026, p < .01) and on goal difficulty (β = .066, SE = .031, p < .01). We also tested the specific mediating effects of procedural constraints and of organizational goal conflicts. We found that the effect of transformational leadership, through procedural constraints, on goal difficulty was statistically significant (β = .076, SE = .027, p < .01) as was the effect of transformational leadership, through organizational goal conflict, on goal specificity (β = .103, SE = .024, p < .01). However, neither the effect of transformational leadership on goal difficulty (through organizational goal conflict) or on goal specificity (through procedural constraints) was statistically significant (see Figure 2). That is, only two of the four suggested pathways were confirmed: procedural constraints mediate the relationship between transformational leadership and goal difficulty, and organizational goal conflicts mediate the relationship between transformational leadership and goal specificity. Our fourth hypothesis is therefore only partly supported by the data.

Table 4 summarizes our results.

Table 4.

Results of Hypotheses Testing.

H1. A transformational leadership style will be positively related to the work motivation of public sector employees	Supported
H2. Goal setting positively mediates the relationship between a transformational leadership style and the work motivation of public sector employees	Supported
H3. The work context of public sector organizations has a negative indirect effect on the work motivation of public sector employees through its negative effect on goal setting.	Partly supported
H4. A transformational leadership style will have an indirect positive effect on goal setting through its effect on the public sector work context.	Partly supported

Conclusions and Discussion

In this article, we have linked a transformational leadership style to the work motivation of public sector employees. In so doing, we add to the framework on work motivation in a public sector work context by Wright (2001, 2004). Overall, we have shown that such a leadership style is indeed related to work motivation. Our analysis shows three linkages. First, there is a direct relationship: a transformational leadership style is associated with higher work motivation. This is probably the result of transformational leaders being able to energize their followers, resulting in a greater work effort.

There are also two indirect linkages between transformational leadership style and work motivation. First, goal setting partly mediates the relationship between transformational leadership and work motivation. As explained in the theoretical section, this linkage can be understood by the fact that the vision set by transformational leaders is cascaded down into more specific and concrete goals. Moreover, a transformational leader will set more challenging goals. Our findings were in accordance with this theoretical argument. It appears that transformational leaders do indeed set more difficult and specific goals and, following the logic of goal-setting theory, this results in a higher work motivation of public employees.

Second, there is also an indirect relationship between a transformational leadership style and the goal-setting process. Based on earlier studies, we suggested that a transformational leadership style would be associated with perceptions of less organizational goal conflict and fewer procedural constraints. Our findings confirm this and show an indirect effect of transformational leadership on goal setting. Our results show that transformational leaders are able to set more challenging (i.e., difficult) goals because they are able to overcome the procedural constraints perceived by their employees. They are also able to set more specific goals as they are able to reduce organizational goal conflicts.

Our analysis also shows that the elements that we have included in our analysis as characteristics of the public sector (organizational goal conflicts and procedural constraints) are indirectly related to work motivation and that goal setting acts as a mediator. However, of the four possible pathways only one is significant: procedural constraints are associated with less difficult goals and thus with lower work motivation of public employees. Although our procedures were not able to test for double mediation, this does suggest that a transformational leadership style also has an indirect effect on work motivation because it reduces procedural constraints which in turn increases goal difficulty.

As outlined in the Introduction, our conceptual framework is based on a combination of findings reported in the literature. In terms of these earlier findings, our study supports the view that a transformational leadership style affects the goal-setting process. Our finding on the indirect effect on goal specificity is in accordance with Moynihan, Wright, and Pandey (2012) and Wright et al. (2012) who found that transformational leaders are able to make goals clearer. Our finding on the indirect effect on goal difficulty through limiting procedural constraints is in accordance with the findings of Walker and Brewer (2009) in that public managers are able to reduce red tape. Given the limitations in our measurements, it is uncertain whether this can be attributed to a genuine reduction in constraints or to a greater willingness of public employees to overcome them (for instance induced by the intellectual stimulation provided by transformational leaders that enable their followers to see the world in a new light).

Secondly, our study builds on Wright’s (2004) study on work motivation in a public context. While our model is more limited in some senses (we did not include some of the variables included in his research such as organizational goal specificity, self-efficacy, and feedback), we have also extended it by adding transformational leadership. On a general level, our findings fit with Wright’s in that, in a public sector context, goal setting affects work motivation, and also mediates the effect of the public sector work context on work motivation. There are however differences in results. The most significant of these is that while Wright found that procedural constraints only have an effect on job goal specificity, we also found an effect on goal difficulty. We cannot explain the differences with certainty. They could stem from differences in model specification (particularly differences in the variables included) or in context (Wright studied New York State employees). One finding in particular seems somewhat surprising: the negative effect of procedural constraints on goal difficulty. Should one not expect constraints to make goals more difficult, as overcoming these constraints makes work more challenging? Our tentative explanation is that these constraints lead to a routinization and formalization of tasks, with the consequence that work goals become less demanding.

Limitations

Several limitations affect the analysis and results. First, our study is limited to knowledge workers in a single Dutch city. Therefore, further research is needed to determine whether the findings can be generalized to other public sector work settings. In this respect, we would encourage other researchers to use similar scales, as this would substantially enhance the possibility for generalization.

Second, although we have included several characteristics that the literature suggests are specific to the public sector, we cannot be certain that Dutch private organizations do really differ from public organizations with respect to organizational goal conflict and red tape. We would again stress that the distinction between public and private organizations should not be exaggerated. Certainly, some private organizations will also have to deal with a high level of red tape and organizational goal conflict.

Third, we have only looked for positive effects of transformational leadership on motivation. As we stressed in the theoretical section, there could also be some negative effects. In this respect, Harrison (1987) warned of a burnout risk and this could negatively affect motivation. Moreover, a transformational leadership style is not universally applicable (van Wart, 2012). Determining situations in which it is, or is not, is therefore another important topic for further research.

Fourth, transformational leadership, although hugely popular, is only one possible leadership style. It is therefore worthwhile to study whether other styles, for instance based on distributed approaches or ethic-based leadership theories (Van Wart, 2012), have similar effects as found in this study.

Fifth, work motivation has many determinants. Transformational leadership is only one of many factors that influence goal setting, and this, in turn, is not the only antecedent of work motivation. Although we do believe we have outlined a relevant mechanism, there are other avenues available if one wants to affect the work motivation of employees.

Sixth, our data are based on a survey and, as such, our analysis uses self-reported perceptual data. This is especially the case in the measurement of transformational leadership style and the public sector work context as, in this research, these are interpreted through the eyes of the employees. Possibly, a different measurement (for instance by interviewing managers) would have resulted in different findings. Moreover, this type of research can result in common-source bias. Given this risk, we conducted a Harman one-factor test (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003) to see if the majority of the variance could be explained by a single factor. A principal component factor analysis with varimax rotation revealed six distinct factors with eigenvalues greater than unity. These six factors together accounted for 62% of the total variance and no factor for more than 24%. Although these results do not preclude the possibility of common-source bias, they do suggest that such a bias is unlikely to have confounded the interpretation of the results. A further consequence of our cross-sectional research design is that it is not possible to draw conclusions about causality as all the variables were measured at the same time. This highlights the need for a longitudinal design in any follow-up study.

Seventh, our model simplified part of the theoretical mechanism. As outlined in the Introduction, many potentially important moderators and mediators are discussed within goal-setting theory, such as the idea that employees have to be committed to their goals and must then have the self-efficacy to achieve them. Further research aiming to extend the model outlined in this article should include such variables.

Finally, building on the previous point, we have viewed goal-setting variables as mediating the relationship between transformational leadership and work motivation. An alternative hypothesis is that goal setting is a moderator rather than a mediator. With this alternative perspective, one might argue that a transformational leadership style will only (or more strongly) affect work motivation when difficult and specific goals are set. Further research should investigate which type of mechanism (mediating or moderating) best fits empirical data.

Footnotes

Appendix: Questionnaire

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Babette Bronkhorst is a PhD student at the Department of Public Administration at Erasmus University Rotterdam. Her research focuses on the relationship between organizational climate, leadership and wellbeing of healh care employees.

Bram Steijn is a full professor HRM in the public sector at the Department of Public Administration at Erasmus University Rotterdam. His research focuses on the relationship between HRM, leadership and motivation of public employees.

Brenda Vermeeren is a PhD student at the Department of Public Administration at Erasmus University Rotterdam. Her research is focused on the relationship between HRM and performance of public organizations.

References

Anderson

Gerbing

(1988). Structural equation modelling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411-423.

Bagozzi

Philips

(1982). Representing and testing organisational theories: A holistic construct. Administrative Science Quarterly, 27, 459-489.

Bass

B. M.

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

Berson

Avolio

B. J.

(2004). Transformational leadership and the dissemination of organizational goals: A case study of a telecommunication firm. The Leadership Quarterly, 15, 625-646.

Bowerman

B. L.

O’Connell

R. T.

(1990). Linear statistical models: And applied approach. Belmont, CA: Duxbury.

Brewer

G. A.

Walker

R. M.

(2010). The Impact on Governmental Performance: An Empirical Analysis. Journal of Public Administration Research and Theory, 20, 233-257.

Burns

J. M.

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

Camilleri

(2006). Towards developing an organisational commitment—Public service motivation model for the maltese public service employees. Public Policy and Administration, 21(1), 63-83.

Carless

S. A.

Wearing

A. J.

Mann

(2000). A short measure of transformational leadership. Journal of Business and Psychology, 14, 389-405.

10.

Davenport

T. H.

(2005). Thinking for a living: How to get better performance and results from knowledge workers. Boston, MA: Harvard Business School.

11.

Densten

I. L.

(2005). The relationship between visioning behaviours of leaders and follower burnout. British Journal of Management, 16, 105-118.

12.

Dvir

Eden

Avolio

B. J.

Shamir

(2002). Impact of transformational leadership on follower development and performance: A field experiment. The Academy of Management Journal, 45, 735-744.

13.

Field

(2005). Discovering statistics using SPSS (2nd ed.). London, England: SAGE.

14.

Harrison

(1987). Harnessing personal energy: How companies can inspire employees. Organizational Dynamics, 16(1), 4-21.

15.

Hayes

A. F.

(2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76, 408-420.

16.

Hox

J. J.

(2003). Non-normal data and bootstrapping in Amos. Utrecht, Netherlands: Utrecht University.

17.

Ilies

Judge

Wagner

(2006). Making sense of motivational leadership: The trail from transformational leaders to motivated followers. Journal of Leadership & Organizational Studies, 13(1), 1-22.

18.

Kirkpatrick

S. A.

Locke

E. A.

(1996). Direct and indirect effects of three core charismatic leadership components on performance and attitudes. Journal of Applied Psychology, 81, 36-51.

19.

Latham

G. P.

(2007). Work motivation. History, theory, research, and practice. Thousand Oaks, CA: SAGE.

20.

Latham

G. P.

Borgogni

Petitta

(2008). Goal setting and performance management in the public sector. International Public Management Journal, 11, 385-403.

21.

Locke

E. A.

Latham

G. P.

(2007). New Developments in and Directions for Goal-Setting Research. European Psychologist, 12, 290-300.

22.

Locke

E. A.

Latham

G. P.

(2002). Building a practically useful theory of goal setting and task motivation. American Psychologist, 57, 705-717.

23.

Locke

E. A.

Latham

G. P.

(1990). A theory of goal setting and task performance. Englewood Cliffs, NJ: Prentice-Hall.

24.

Masi

R. J.

Cooke

R. A.

(2000). Effects of transformational leadership on subordinate motivation, empowering norms, and organizational productivity. International Journal of Organizational Analysis, 8, 16-47.

25.

Moynihan

D. P.

Pandey

S. K.

Wright

B. E.

(2012). Setting the table: How transformational leadership fosters performance information use. Journal of Public Administration Research and Theory, 22, 143-164.

26.

Moynihan

D. P.

Wright

B. E.

Pandey

S. K.

(2012). Working within constraints: Can transformational leaders alter the experience of red tape? International Public Management Journal, 15, 315-336.

27.

Perry

J. L.

Mesch

Paarlberg

(2006). Motivating employees in a new governance era: The performance paradigm revisited. Public Administration Review, 66, 505-514.

28.

Perry

J. L.

Porter

L. W.

(1982). Factors affecting the context for motivation in public organizations. The Academy of Management Review, 7, 89-98.

29.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903.

30.

Podsakoff

P. M.

MacKenzie

S. B.

Moorman

R. H.

Fetter

(1990). Transformational leader behaviors and their effects on followers’ trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly, 1, 107-142.

31.

Rainey

H. G.

(2009). Understanding and managing public organizations (4th ed.). San Francisco, CA: Wiley/Jossey-Bass.

32.

Rainey

H. G.

Pandey

Bozeman

(1995). Research note: Public and private managers’ perceptions of red tape. Public Administration Review, 55, 567-574.

33.

Shamir

House

R. J.

Arthur

M. B.

(1993). The motivational effects of charismatic leadership: A self- concept based theory. Organization Science, 4, 577-594.

34.

Stephens

C. U.

D’Intino

Victor

(1995). The moral quandary of transformational leadership. Change for whom? Research in Organizational Change and Development, 8, 123-143.

35.

Van Wart

(2012). Leadership in public organizations. An introduction (2nd ed.). New York, NY: Sharpe.

36.

Van Wart

(2013). Administrative Leadership Theory: A reassessment after 10 years. Public Administration, 91, 521-543.

37.

Walker

R. M.

Brewer

G. A.

(2009). Can public managers reduce red tape? The role of internal management in overcoming external constraints. Policy & Politics, 37, 255-272.

38.

Wang

I. -S.

Courtright

S. H.

Colbert

A. E.

(2011). Transformational leadership and performance across criteria and levels: A meta-analytic review of 25 years of research. Group & Organization Management, 36, 223-270.

39.

Whittington

J. L.

Goodwin

V. L.

Murray

(2004). Transformational leadership, goal difficulty, and job design: Independent and interactive effects on employee outcomes. The Leadership Quarterly, 15, 593-606.

40.

Wright

B. E.

(2004). The role of work context in work motivation: A public sector application of goal and social cognitive theories. Journal of Public Administration Research and Theory, 14, 59-78.

41.

Wright

B. E.

(2001). Public sector work motivation: A review of the current literature and a revised conceptual model. Journal of Public Administration Research and Theory, 11, 559-586.

42.

Wright

B. E.

Moynihan

D. P.

Pandey

S. K.

(2012). Pulling the levers: Transformational leadership, public service motivation, and mission valence. Public Administration Review, 72, 206-215.

43.

Wright

B. E.

Pandey

S. K.

(2010). Transformational leadership in the public sector: Does structure matter? Journal of Public Administration Research and Theory, 20, 75-89.

44.

Yukl

(1999). An evaluation of conceptual weaknesses in transformational and charismatic leadership theories. Leadership Quarterly, 10, 285-305.