Empowerment in the Public Sector: Testing the Influence of Goal Orientation

Empowerment has emerged as an important new issue in the public sector organization setting in the wake of mainstream new public management (NPM). Nevertheless, few studies in this frame have combined structural (managerial) and psychological (individual) approaches in an integrative study of empowerment. There is also a need to examine the moderating variables involved in this relationship, as well as to extend research on work motivation in public management. This study explores the effect of structural empowerment on psychological empowerment, and it also draws on goal orientation (GO) theory to examine the moderating role of employees’ GO in this link. The model is tested on a sample of 521 Spanish local authority employees. The results do not confirm the direct link between structural and psychological empowerment, but show that learning GO has considerable moderating power in this relationship, and its interaction with structural empowerment affects employees’ psychological empowerment levels.


Introduction
Empowerment is a key element within new public management (NPM; Fernandez & Moldogaziev, 2013), the new orientation in the public sector that aims to provide a better service for citizens (Nicholson-Crotty, Nicholson-Crotty, & Fernandez, 2017).
1 Universitat Jaume I, Castelló de la Plana, Spain In many countries, the public sector has undergone reforms attempting to change the structure, way of management, and even the culture of public administration (Üstüner & Coşkun, 2004;Worthington & Dollery, 2001), as well as the management of people (Truss, 2013). Peters and Savoie (1996) describe how the current form of government proposes more decentralized organizations and more empowered employees at all levels of the hierarchy. Thus, empowerment has reemerged as a component of significant government reforms around the world (Brewer & Kellough, 2016). It is closely linked to the drive to enhance organizational effectiveness through the wise use of human resources (Siegall & Gardner, 2000), and in this sense, many studies have found positive effects of implementing empowerment (see, for example, Maynard, Gilson, & Mathieu, 2012). Overall, as Dimitriades (2005) notes, it "is essential to the constant change and learning that characterize today's global organizational environment" (p. 80).
Nevertheless, and despite the numerous studies on the subject and the importance attributed to empowerment, it "is a tricky concept to grasp" (Matthews, Diaz, & Cole, 2003, p. 315), and no single definition has been agreed on (Petter, Byrnes, Choi, Fegan, & Miller, 2002). Empowerment has, therefore, been examined from several separate perspectives (H.-F. Chen & Chen, 2008;Dimitriades & Maroudas, 2007;Mathieu, Gilson, & Ruddy, 2006), although two main approaches can be distinguished: structural and psychological. The structural perspective views empowerment as a set of practices, conditions, policies, and structures that enable the transfer of power and authority from higher levels of the organization to lower levels, increasing access to information and resources (Bowen & Lawler, 1992;Mathieu et al., 2006). It is "a managerial-initiated, socio-structural phenomenon" (Biron & Bamberger, 2010, p. 164); hence, it refers to empowering managerial practices. Overall, it constitutes the "macro" approach to empowerment (Biron & Bamberger, 2010). The psychological perspective, in contrast, is based on Bandura's (1977) work on self-efficacy (Maynard et al., 2012), and defines empowerment as "a process of enhancing feelings of self-efficacy among organizational members through the identification of conditions that foster powerlessness and through their removal by both formal organizational practices and informal techniques of providing efficacy information" (Conger & Kanungo, 1988, p. 474). Following this idea, Thomas and Velthouse (1990) describe empowerment "as increased intrinsic task motivation" (p. 666). This perspective considers the individual employee level, representing the "micro" approach to empowerment (Biron & Bamberger, 2010), Many authors (e.g., Laschinger, Finegan, & Shamian, 2001;Mathieu et al., 2006;Maynard et al., 2012;Menon, 2001) argue that these two perspectives must be linked to arrive at a broader understanding of the empowerment process. The few recent studies that have dealt with this issue have positioned structural empowerment practices as antecedents of psychological empowerment (e.g., Hempel, Zhang, & Han, 2012;Seibert, Silver, & Randolph, 2004). However, public sector scholars have paid little attention to the study of empowerment in general, calling for more studies in this field (Carless, 2004), and pointing to the lack of research on the relationship between structural and psychological empowerment (Cho & Faerman, 2010).
Empirical contributions on this relationship are, therefore, important, particularly as empowerment is a very useful tool with which to face the challenges of the new public context. As Cho and Faerman (2010) note, to benefit from the effectiveness of the new techniques, leaders in government organizations should pay attention to both structural and psychological empowerment.
Furthermore, there is also a need to examine how structural empowerment can interact with other variables to achieve psychological empowerment (Maynard et al., 2012). The literature review shows that the effect of moderator variables in this relationship has so far been neglected. In this vein, Maynard et al. (2012) propose the study of the effect of individual orientations, such as goal orientation (GO; for example, Dweck, 1986). Vandewalle (1997) understands the concept of GO as "an individual disposition toward developing or validating one's ability in achievement settings" (p. 995). It is, therefore, a personality variable and an individual difference (VandeWalle, Brown, Cron, & Slocum, 1999). Drawing from this view, we focus on the study of employees' GO as a moderating variable in the structural and psychological empowerment relationship. Employees' GO may be a key factor at work and in workforce selection processes, as it explains the differences in personal work interests and behaviors, and in receptiveness to and interpretations of professional development activities (Joo, Park, & Lim, 2016;Vandewalle, 1997). Structural empowerment practices could, therefore, affect psychological empowerment feelings in very different ways, depending on employees' GO, because they perceive and respond differently to such practices, which is likely to yield different psychological empowerment results.
The objective of this study is to conceptually develop and empirically test a model proposing that structural empowerment is related to psychological empowerment in employees, and that this link is different depending on their GOs. Our study responds to Petter et al.'s (2002) call for more academic attention to how public organizations influence their employees, providing more successful and integral research on empowerment, and to Maynard et al.'s (2012) call for research on the moderator factors that may influence the structural and psychological empowerment relationship. The study makes two main contributions to the literature. First, by examining the structural-psychological empowerment relationship, we extend the scant empowerment literature in public management. As structural empowerment has largely been emphasized as an administrative strategy to increase work-related outcomes, rarely considering the psychological empowerment approach (Cho & Faerman, 2010), evidence on psychological empowerment in public organizations is scarce (Dimitriades & Maroudas, 2007). Furthermore, research on the structural-psychological empowerment link in the public sector has mainly looked at the field of health, particularly nursing (Laschinger et al., 2001;Laschinger, Finegan, Shamian, & Wilk, 2004). Cho and Faerman (2010) carried out one of the few studies on the structural and psychological empowerment relationship in the public sector (other than the health services), finding that structural empowerment affects extra-role performance through psychological empowerment in a sample of public employees working in the city of Seoul (Korea). Therefore, our work aims to provide new empirical evidence about empowerment in the public sector by addressing the link between its two approaches in the local authority context. In doing so, we follow Lawler's (1992, 1995) structural empowerment conceptualization and Spreitzer's (1995) conceptualization of psychological empowerment, and relate them to the public administration framework.
Second, we examine the moderating effects of employees' GO levels in this relationship. We aim to add to the empowerment literature by extending the limited understanding of possible moderator variables. Furthermore, to the best of our knowledge, public administration research has not examined the GO concept, leaving a significant research gap. Likewise, more generally, only one empirical study (Joo et al., 2016) has dealt with the influence of GO on psychological empowerment, although it focuses on only one of its dimensions (learning GO). Given that public sector human resources are the main input in delivering services, examining people's motivations to achieve in their work context is an issue of great interest, and in this study, we respond to the need to consider motivational aspects in the public sector (Perry, 2010;Vandenabeele, Depré, Hondeghem, & Yan, 2005). Local governments are a permanent workplace for many employees (Barba & Serrano, 2015), and are the level of government closest to citizens. Therefore, employees' GO, and its interaction with structural empowerment practices emanating from new managerial techniques and legal rules (NPM, made evident through EBEP, 2007, Estatuto Básico del Empleado Público, in English, the Basic Statute of the Public Employee), as well as its corresponding effects on employees' psychological state (here, psychological empowerment), are important issues to examine. To address this, we review arguments from GO theories and previous related contributions.
In light of the above, we consider that the present study can shed more light on the question of empowerment in the public administration, local governments in this case, and can also help to bridge the gap in the literature on the moderating variables in empowerment research and the motivational issue in the public sector. The rest of the article is organized as follows: the next section provides a review of the literature on empowerment, GO, and the public context, and presents the study hypotheses. The data and methods used are then explained, followed by the results of the analysis. Finally, we present the discussion, with the study's implications, limitations, and conclusions.

Theoretical Framework
We consider Bowen and Lawler's vision of empowerment as the most suitable for the current study, following previous contributions (e.g., Fernandez & Moldogaziev, 2011Fernandez, Resh, Moldogaziev, & Oberfield, 2015) that have also dealt with empowerment in the public sector. As Fernandez and Moldogaziev (2011) state, Lawler (1992, 1995) develop the most widely known definition of structural empowerment, analyzing the use of empowerment practices in private service firms, and defining it as the "approach to service delivery" (Bowen & Lawler, 1992, p. 32). According to Lawler (1992, 1995), sharing power and authority with frontline employees is a key element of an empowering approach to management, but it is not sufficient to achieve the benefits of (structural) empowerment. These authors identify four practices that need to be considered to achieve a notable effect on performance: (a) sharing power to make decisions, (b) redistributing information about the organization's performance, (c) providing knowledge that enables employees to understand and contribute to organizational performance, and (d) offering rewards based on the organization's performance (1992). Empowerment exists when all these practices are in place. Spreitzer's (1995) concept of psychological empowerment has received a great deal of international recognition and is followed by the majority of authors. Following Thomas and Velthouse (1990), Spreitzer (1995) describes psychological empowerment "as a motivational construct manifested in four cognitions: meaning, competence, self-determination, and impact" (p. 1444). Meaning is the judgment of the fit between the ideals and standards of the employee and his or her work goal or purpose. The second factor, competence, is the individual's belief in his or her capability to carry out work tasks with skill. Third, self-determination or autonomy is related to the feeling that one is able and allowed to initiate and continue work actions. Finally, impact is the degree to which individuals feel they can influence outcomes at work (Spreitzer, 1995;Thomas & Velthouse, 1990).
Regarding our context of study, the components or dimensions of psychological empowerment (meaning, competence, self-determination, and impact) are essential factors in local government human resources management and help employees to face the challenges of the NPM environment. For public employees, feelings of empowerment are highly relevant to their work routine, as they usually want to serve the public interest, and are more motivated by the intrinsic compensation they receive from accomplishing tasks (Houston, 2000). Optimal service is determined by employees feeling they exercise control over how they work to achieve their job objectives (Cuenca, 2010). This helps them to perceive that the activities and tasks that they carry out have an effect on the general results (Taylor, 2013). In addition, the modern local government context implies that employees must feel that they are making decisions and developing activities autonomously, in direct contrast to concentrated levels of authority (Vandenabeele et al., 2005).

Structural and Psychological Empowerment Link
Within the empowerment framework, research initially focused on structural empowerment (e.g., Bowen & Lawler, 1992), analyzing issues such as distribution of power and decision making, training or planning, and control systems. The emphasis then shifted to the psychological state of empowerment, but neglected the structures and practices that actually stimulate empowerment (Hempel et al., 2012). Wisely, some researchers (e.g., Menon, 2001) have highlighted the need to examine both types of empowerment in combination. Thomas and Velthouse's (1990) vision of empowerment in the workplace (which spawned Spreitzer's definition of psychological empowerment) held that psychological empowerment is influenced by the work environment, and therefore, empowerment practices (structural empowerment) help employees to develop and create an active and optimistic orientation to their job and to feel that they can carry out their work activities appropriately. Therefore, empowering practices create an empowered state of mind in employees that includes control over what happens on the job, autonomy in doing the job and acting naturally with customers, and consciousness of the work environment, among others (Bowen & Lawler, 1995). All these feelings are closely related to Spreitzer's psychological empowerment dimensions.
Later studies combining the two empowerment approaches (e.g., Hempel et al., 2012;Laschinger et al., 2004;Seibert et al., 2004) positioned structural empowerment as a necessary antecedent to psychological empowerment (Maynard et al., 2012). One of the first researchers to propose this link was Spreitzer (1995Spreitzer ( , 1996, who found that low role ambiguity, sociopolitical support, access to information, and participative climate generate potential for (psychological) empowerment in the workplace. Other authors who have studied this relationship in a wide variety of service firms (Wallace, Johnson, Mathe, & Paul, 2011) affirm that empowering practices such as autonomy, reward system, and participative decision making have beneficial effects on the psychological empowerment level of the employees.
Despite its importance and the empirical evidence found, empowerment, especially psychological empowerment, has rarely been studied in the public sector context (Cho & Faerman, 2010;Dimitriades & Maroudas, 2007). Cho and Faerman (2010) conduct one of the few studies examining the structural and psychological empowerment link in an area of the public sector other than health services. Using a sample of public employees from the city of Seoul (Korea), they found that structural empowerment (participative decision making, delegation and feedback) affects extra-role performance through psychological empowerment. Self-determination theory (Deci & Ryan, 1980;Deci et al., 2001) can provide theoretical support for such linkages. This theory holds that everybody has three basic psychological needs: competence, autonomy, and relatedness. Employees will feel psychological well-being and motivation when they perceive these needs are satisfied in their organizations. Structural empowerment provides employees with possibilities to enhance their abilities and knowledge through training and information programs, to make decisions on organizational and job matters, and to relate to their peers by allowing them to become involved in decisionmaking programs. Thus, structural empowerment interventions would help to meet employees' basic needs, enhancing their intrinsic motivation and well-being, and therefore, their feelings of psychological empowerment.
Keeping in mind the above, it is clear that elements of structural empowerment, such as sharing information or granting decision-making power, are key ways to boost employees' feelings of psychological empowerment. Although most studies on the link between structural and psychological empowerment have focused on health service organizations, or other service sector firms, in light of the studies reviewed, the legislative tendencies, and the local administration context, structural empowerment is likely to be valuable and have considerable repercussions on the psychological empowerment of local government employees. Therefore, we propose that Hypothesis 1 (H1): There is a positive relationship between structural empowerment and psychological empowerment.

Employees' GO
GO is a motivational concept (Cellar et al., 2011;DeShon & Gillespie, 2005;Lee, Tinsley, & Bobko, 2003) as it affects the employee's level of motivation (Button, Mathieu, & Zajac, 1996;Spinath & Steinmayr, 2012). The growing research on motivation in public sector organizations (Vandenabeele, 2013) shows similarities (e.g., Buelens & Van den Broeck, 2007) as well as differences (e.g., Houston, 2000) between public and private employees. Some types of goals have been studied in the public sector, such as goal content or goal commitment (e.g., Wright, 2001Wright, , 2007. However, as Wright (2001) states, there is still a lack of attention to work motivation in the public sector, and a better understanding of this question is crucial to improve the efficiency and effectiveness of public organizations. After all, human resources make the difference, and their motivation, and therefore their GO, should be regarded as essential to public sector organizations (Vandenabeele, 2013).
There is a certain consensus about taking a three-dimensional approach to the conceptualization of GO (e.g., Brett & VandeWalle, 1999;Elliot & Church, 1997;Vandewalle, 1997), which has been most commonly used in recent years (Phan, 2009). This conceptualization of GO comprises the following three factors: • • Learning goal orientation (LGO): directed to increase competence by acquiring new skills and mastering new circumstances (Dweck, 1986), and involves rising to challenges and learning in a certain performance area (G. Chen, Gully, Whiteman, & Kilcullen, 2000). • • Prove-performance goal orientation (PPGO): desire to demonstrate one's own aptitude and to be positively judged on it (Vandewalle, 1997), with an intrinsic enthusiasm to compete against others (Dietz, van Knippenberg, Hirst, & Restubog, 2015). • • Avoid-performance goal orientation (APGO): desire to avoid disapproval of one's capability and undesirable judgments on it (Vandewalle, 1997); characterized by avoidance of inadequate performance and its negative results (Martin, Marsh, & Debus, 2001).

The Moderating Role of GO on the Relationship Between Structural Empowerment and Psychological Empowerment
GO theory (Dweck, 1975;Pintrich, 2000;Vandewalle, 1997) strives to determine what leads people to respond differently to achievement settings (Rusk & Rothbaum, 2010). It can, therefore, be useful to explain the impact of the GO dimensions on the structural-psychological empowerment link within the public administration framework. As Vandewalle (1997) states, every employee is different, and differences in their work interests and actions are, therefore, to be expected. The explanation for these differences has to do with GOs (Vandewalle, 1997). In line with this argument, we analyze the different kinds of GOs and their moderating power. Individuals with high LGO are inclined to think that their capabilities are malleable and can be extended (Brett & Atwater, 2001). Thus, people with high LGO approach tasks with the aim of developing their capabilities (Lee et al., 2003), and take pains to understand new information (Wolters, 2003). It makes sense, therefore, to assume that LGO will be positively associated with willingness to take part in training activities (Button et al., 1996), as these individuals believe their ability is malleable and can be improved (Brett & VandeWalle, 1999). This idea is also supported by Towler and Dipboye (2001), whose empirical study showed that LGO positively moderates the effects of training: the high learning goal-oriented trainees were more motivated to learn and responded more positively to the trainer and the lecture. Likewise, as individuals with strong LGO are more motivated to use learning processes when facing problems and obstacles, they develop expertise more easily and create strategies to become successful (Gong, Huang, & Farh, 2009), which can strengthen the effects of structural empowerment interventions on, for instance, their feelings of competence and impact. Furthermore, people with this GO perceive negative feedback as valuable information on how to develop mastery (Dweck, 1986), and perform better in challenging tasks, as demonstrated by Preenen, van Vianen, and de Pater (2014), who found a positive relationship between employees' masteryapproach orientation and performance of challenging tasks on a sample of workers. According to Coad's (1999) study in an accounting environment, accountants with LGO are more likely to become involved in the management decision process. Therefore, it is plausible that in the context of structural empowerment implementation employees with high LGO will experience higher levels of psychological empowerment. Structural empowerment interventions will have a higher positive association with the development and enhancement of employees' psychological empowerment because they provide empowerment practices designed to increase employees' feelings of control over their work and over their skills to perform activities, which can be more easily achieved when employees are willing to be trained, to learn, to improve their skills, to meet challenges, and to become familiar with new information. In the specific context of NPM, in which public employees have to learn and master new demands and tasks to provide better services, their levels of LGO combined with structural empowerment practices could play a pivotal role in their feelings of empowerment. Thus, the review of the literature suggests that the higher the level of LGO, the stronger and more positive the link between structural empowerment practices and psychological empowerment will be.
The empirical evidence related to PPGO is inconclusive (Midgley, Kaplan, & Middleton, 2001) due to mixed findings (e.g., Elliot & Moller, 2003;Payne, Youngcourt, & Beaubien, 2007) depending on whether the more traditional or the revised perspective of GO is considered. The more traditional perspective (e.g., Ames, 1992) holds that any concern with performance could have detrimental effects on participation in a task due to the distraction brought about by comparisons with others or negative self-evaluation (Pintrich, 2000). However, according to the most recent perspective, the revised goal theory (Pintrich, 2000), performance goals may be a useful tool as they offer an external orientation with which to assess performance, and they are a guide for understanding feedback (Butler & Winne, 1995).
Drawing on this latest perspective, and given that individuals with high PPGO attempt to show their ability by trying to appear better than others (Brett & VandeWalle, 1999), we assume that they will see opportunities to demonstrate their ability when they are given the chance to take decisions by themselves, are rewarded for their productivity, or receive training, thus enhancing the effects of structural empowerment on psychological empowerment. Therefore, and from the limited literature on this issue (Joo & Park, 2010), which postulates a more adaptive role for prove performance goals (Pintrich, 2000), we consider that the higher the level of PPGO, the more likely structural empowerment practices (e.g., knowledge and training, power to make decision) are to have a positive relationship with psychological empowerment. The latest reforms in the public sphere have attempted to improve human resource management and to use public resources more efficiently by implementing methods such as performance appraisal (Barba & Serrano, 2015). High-PPGO employees' levels of psychological empowerment could rise in this new situation, in which their individual performance and comparison with other workmates is revealed, an aspect they pursue and enjoy.
Finally, research on APGO has uncovered its relationship with maladaptive patterns and responses (Chadwick & Raver, 2015;Pintrich, 2000). As individuals with this GO seek to avoid criticism of their competences and negative judgments from others (Vandewalle, 1997), we expect that structural empowerment practices, such as delegating decision-making power, will have a lower association with psychological empowerment levels when APGO is high. The reason for this assumption is that these individuals feel threatened by the achievement context (Elliot & Harackiewicz, 1996), and we consider that the possibility of taking decisions by themselves, being rewarded for their performance, or participating in training activities can be interpreted as a threat, due to fear of failure (Spinath & Steinmayr, 2012). In addition, APGO triggers a process that complicates ideal task engagement (Elliot & Harackiewicz, 1996). Hence, feelings of psychological empowerment derived from structural empowerment interventions are likely to be lower among individuals with high APGO, as mechanisms related to APGO may erode the positive effects of practices such as training, information, rewards, or decision making. Moreover, this GO predicts self-handicapping (Midgley & Urdan, 2001), so the lower the degree of avoid-performance GO, the stronger and more positive the association of structural empowerment practices, such as knowledge and training, and decision-making power, with psychological empowerment. In contrast, the self-handicapping feelings derived from high APGO will hamper the emergence of positive feelings of empowerment, as high-APGO individuals will perceive that they cannot face structural empowerment practices that entail carrying out new activities and meeting new challenges, such as training in new fields or suggesting ideas for the work routine. For high-APGO public employees, the possibility of failure after a performance appraisal could reduce feelings of psychological empowerment.
Thus, from the findings of previous contributions and the theoretical arguments exposed above, we put forward the following hypotheses:
Our conceptual model is represented in Figure 1.

Sample and Data Collection
In the context of Spain's new legal rules (EBEP, 2007), Spanish local authorities were considered as a suitable frame to test the relationship between the two empowerment approaches. Spanish local governments have undergone significant modifications in recent years (Suárez Barraza & Ysa, 2011). The EBEP aims to enhance participation and improve human resource management through principles and actions related to structural empowerment (e.g., training, performance-related bonuses, and the importance of producing understandable and meaningful structures) for local government employees (Cuenca, 2010) that also promote their psychological empowerment. Citizens today receive highly professional and diverse services (Vermeeren, Kuipers, & Steijn, 2011), which reflects the importance of the human factor in current administrative institutions (Giménez & Prior, 2007;Kim & Wright, 2007).
The fieldwork for this study took place in Spanish municipalities of more than 20,000 inhabitants. We focused on large municipalities as they implement more strategic management techniques (Poister & Streib, 2005), including structural empowerment techniques (Seibert, Wang, & Courtright, 2011). Following previous contributions (e.g., Barba & Serrano, 2015;Rosa, Morote, & Colomina, 2013) we used the number of inhabitants as an index, which yielded a population of 399 municipalities taken from the Federación Española de Municipios y Provincias (FEMP, Spanish federation of municipalities and provinces) database.
Two surveys were prepared: one for local government managers (e.g., human resource manager, clerk), and a second for other public workers. Collecting data from two different sources limits problems associated with common method variance (Collins & Smith, 2006;Gardner, Wright, & Moynihan, 2011). The questionnaires were then uploaded onto an Internet survey application and pretested with four local government managers and other employees to assess and improve the measures. Improvements were then made to the questionnaires in accordance with the outcomes of the pretest. The data collection procedure involved first telephoning the city councils and identifying the human resource managers or the person in charge of personnel management (sometimes the clerk or the auditor) to describe the study and request their email addresses. Emails containing links to the two surveys were sent to these managers. They were asked to respond to the managers' questionnaire and were invited to email the employees' questionnaire at random to a minimum of three employees. A follow-up telephone call was made to increase the response rate, as recommended in the literature (Dillman, Smyth, & Christian, 2009).
We removed city councils with fewer than three employee responses, which resulted in a sample of 109 manager questionnaires and 521 employee questionnaires, representing 27.32% of the total population. The sample error for the local governments sample was ±8.09 at 5% significance level. In addition, the presence of nonresponse bias was examined by testing for differences between responses that were included in the study (local governments with responses from both managers and employees) and responses from local governments not included because not all the required questionnaires were completed (e.g., only one employee filled in the questionnaire; Whitehead, Groothuis, & Blomquist, 1993) through a t test on the scores. No significant differences (p < .05) were found in any of the main or control variables; no evidence of nonresponse bias was, therefore, detected in the study. The average number of employees (excluding the human resource manager) participating in each city council was 4.78. The respondents surveyed belonged to a wide range of local authority administrative services, including human resource management, social services, accounting, tourism, and town planning. The majority of the participants were women (62.2%) and nonsupervisors (67.4%). Their average age was 46.71 years (SD = 7.58), and they reported an average tenure in their organization of 17.25 years (SD = 9.56).

Measures
The appendix section provides a complete description of the instruments used in our study. The scales were constructed originally in English. As recommended (e.g., Ghauri, Gronhaug, & Kristianslund, 1995), we followed standard translation and back-translation procedures to translate the measures into Spanish.
Structural empowerment was measured using 22 items adapted from Lawler, Mohrman, and Benson's (2001) scale. Local authority managers responded to items on the four dimensions-decision-making power, information sharing, rewards, and knowledge and training-on a 7-point scale ranging from 1 (no employees) to 7 (all employees). Following previous studies (e.g., Datta, Guthrie, & Wright, 2005), a single indicator was calculated by taking the mean of the four subscales generated from the questionnaire items. The internal reliability coefficient (Cronbach's α) for this composite measure was .93.
We used Spreitzer's (1995) psychological empowerment scale to measure psychological empowerment; this is the most widely used scale in the literature. It consists of four subscales, one for each dimension of empowerment (meaning, competence, selfdetermination, and impact). The 12 items are answered on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). The employees' responses were averaged to create an overall psychological empowerment index following previous literature (e.g., Seibert et al., 2004;Taylor, 2013). Taken together, the items showed high reliability (α = .86).
Finally, in line with previous literature (e.g., Jong & Ford, 2016), we controlled for tenure in the organization (years working in the city council), gender (male = 0 or female = 1), job position (supervisor = 1 or nonsupervisor = 0), age (years), size of the local government (log-transformed number of employees), given that large municipalities put into practice more strategic management techniques (Poister & Streib, 2005), and the strategic integration of HRM in the overall local authority strategy. This final aspect was measured by asking whether the human resource manager or the person in charge of human resource issues participates in city council meetings in which important questions are debated (yes = 1; no = 0). The rationale behind the inclusion of this variable is that the strategic importance of the human resource question in an organization can affect its results (e.g., Huang, 2000).

Tests of the Measurement Model
Confirmatory factor analyses CFA) for each scale were conducted to examine the dimensionality of GOs (LGO, PPGO, and APGO), psychological empowerment, and structural empowerment. First, a CFA was performed to assess the fit of a four-factor model for the 12 items of psychological empowerment, providing a good fit, S-B χ² = 107.6541, df = 48; BBNNFI = 0.969; comparative fit index (CFI) = 0.977; root mean square error approximation (RMSEA) = 0.049. Then, a second-order factor CFA was estimated in which the first-order constructs of meaning, competence, self-determination, and impact were each loaded onto a single second-order latent construct representing psychological empowerment. The higher order model also demonstrated an adequate fit to the data (S-B χ² = 140.2558, df = 50; BBNNFI = 0.955; CFI = 0.966; RMSEA = 0.059), so the model supports the higher order factor structure of psychological empowerment. Regarding GO, a CFA of the 13 items loading onto their corresponding factors (LGO, PPGO, and APGO) indicated a good fit to the data for the three-factor model (S-B χ² =208.0820, df = 62; BBNNFI = 0.900; CFI = 0.919; RMSEA = 0.067), thereby supporting past findings (e.g., Brett & VandeWalle, 1999;VandeWalle, Cron, & Slocum, 2001). Finally, the confirmatory factor analysis results showed an adequate fit of a four-factor model for  Table 1 shows means, standard deviations, and correlation coefficients for the study variables. Contrary to expectations, structural empowerment and psychological empowerment are not correlated. Regarding GO, both LGO and PPGO are positive and significantly correlated with psychological empowerment (r = .41, p < .01, and r = .23, p < .01, respectively), but no significant link is found between psychological empowerment and APGO.

Tests of Research Hypotheses
Hierarchical regression analysis was conducted to test the research hypotheses using SPSS 22 computer software. As recommended in the literature (Aiken & West, 1991), we centered the variables to perform the analysis of the interaction between structural empowerment, LGO, PPGO, and APGO. Table 2 shows the results of the hierarchical regression procedure, first introducing the main effects and the second-order interaction effects as a third-order interaction is posited. To verify that ordinary least squares (OLS) was the appropriate estimator to perform the analyses and avoid possible problems of model fit or misspecification, we tested the conditions of the Gauss-Markov Theorem (Wooldridge, 2010). Thus, linearity in parameters and the normality of data were confirmed; random sampling was the method used to collect data; VIF (variance inflation factor) values were below 2.5 in all the variables (control and main variables), showing no perfect collinearity; the Durbin-Watson statistic was 2.07, showing the independence of the error terms; and, finally, we assumed equality of variances  given the absence of any pattern in the scatterplot of the estimates (values predicted by the model) plotted against the standardized residuals. These results allowed us to infer that OLS represented the best linear unbiased estimator (BLUE; Wooldridge, 2010). Hypothesis 1 proposed a positive direct link between structural empowerment and psychological empowerment. Contrary to our expectations, as displayed in Table 2, this link was not confirmed, as the regression coefficients are not significant for either Model 2 or for the full Model 3. Hypotheses 2a, 2b, and 2c proposed that the levels of LGO, PPGO, and APGO of local authority employees, respectively, moderate the effects of structural empowerment of psychological empowerment. As shown in Model 3 of Table 2, only Hypothesis 2a is confirmed, as the interaction effect of structural empowerment and LGO was significant and positive (β = .12, t = 3.10, p < .01).
To better interpret the above results, Aiken and West (1991) recommend calculating and graphically representing the simple regression line of the principal variable (here, structural empowerment) on the dependent variable (here, psychological empowerment), according to the moderator (here, LGO). To do so, LGO was dichotomized based on the standard deviation above (high) and below (low) its mean value. The graphic representation (Figure 2) shows the psychological empowerment values calculated from the higher and lower values of structural empowerment, defined respectively by a standard deviation above (0.85) and below (−0.85) its mean value. Thus, a LGO = learning goal orientation; PPGO = prove-performance goal orientation; APGO = avoid-performance goal orientation; HRM = human resources management. *p < .10. **p < .05. ***p < .01.
comparative analysis was made between two scenarios: high LGO and low LGO. Note that when the levels of LGO are high, the link between structural empowerment and psychological empowerment is positive and significant (β = .11, t = 1.93, p < .10). However, when employees have low levels of LGO, more local government structural empowerment practices result in lower levels of employees' psychological empowerment, as the relationship is significant and negative (β = −.13, t = −2.27, p < .05).

Supplementary Analyses
To further examine the relationships between structural empowerment, psychological empowerment, and the moderating role of GO dimensions, we repeated the hierarchical regression analysis by estimating four models, one for each of the four dimensions of structural empowerment (information, training and knowledge, rewards, and decision-making power) instead of using the single index. The results show that the models with training and knowledge, and decision-making power dimensions maintain the same significant relationship as that considering the overall construct of structural empowerment; that is, the interaction of LGO and each of these two dimensions presented a significant direct influence on psychological empowerment. In contrast, by themselves the information and rewards dimensions do not show any significant link, neither a direct link nor with the interactions with each one of the GO dimensions. These findings suggest that employees' psychological empowerment levels are only fostered when LGO is combined with providing training for employees and empowering them to make decisions.

Contribution and Theoretical and Practical Implications
The literature review evidences the importance of empowerment in the current human resources management field. Furthermore, given the implementation of NPM reforms in the public sector, local governments offer an illustrative context for this research because of the growing significance of empowerment. Structural empowerment practices are thought to have beneficial effects for employees, including psychological empowerment. In addition, we examined employees' GOs in response to calls for research on individual variables that may moderate this relationship, and on motivation issues in the public sector context. GO theory offers insights into how to link and examine the relationships among the variables of interest. Our findings partially support our predictions. Contrary to our expectations, the structural empowerment practices did not explain the differences in psychological empowerment levels of employees.
LGO emerges as a powerful moderator of this relationship, although PPGO and APGO had no effect. A theoretical discussion of some of the specific contributions of these results now follows. First, our study did not find a significant link between structural empowerment and psychological empowerment, in contrast to previous studies (e.g., Cho & Faerman, 2010;Wallace et al., 2011). There are several possible explanations for these unexpected findings. The first of these is that public sector employees may construct psychological empowerment in a different way to employees of private organizations. The intrinsic rewards that public sector employees receive from performing a task are the most important motivating factor (Houston, 2000); it, therefore, seems that their emotional states and internal motivations are more important, and do not depend on external circumstances as much as those of private sector workers. Arguments from job crafting theory may support such reasoning. This theory holds that "employees craft their jobs by changing cognitive, task, and/or relational boundaries to shape interactions and relationships with others at work" (Wrzesniewski & Dutton, 2001, p. 179). Public employees may create their own psychological empowerment feelings, and may be less affected by external interventions such as structural empowerment. Thus, as Nielsen (2013) states, they are the main actors involved in shaping the results of the organizational interventions. Employees, therefore, have a direct influence on how they feel about their job; they affect their own meaning, competence, impact, and states of autonomy, rather than being affected by structural empowerment initiatives. In this line of argument, previous studies on the link between employees' individual characteristics and the structure of their psychological empowerment have shown that, for example, high levels of need for achievement (Hon & Rensvold, 2006) and positive self-evaluation traits (Seibert et al., 2011) are closely associated with psychological empowerment. In sum, employees' personal features and their own dominance over their mental frameworks and feelings should not be ignored in the public sector context.
Another possible explanation for our unexpected findings on the nonsignificant effects of structural empowerment on psychological empowerment may be related to publication bias. Significant relationships predominate in the published literature (Sutton, Duval, Tweedie, Abrams, & Jones, 2000). Therefore, studies that do not support the structural-psychological empowerment relationship are less likely to be published, which may explain the prevalence of studies reporting this positive and significant link, and the lack of research showing a nonsignificant relationship.
Second, our hypotheses on the moderating variables have been partially confirmed. High levels of LGO strengthen the linkage between structural and psychological empowerment notably, such that it becomes significant and positive. Therefore, the interaction of structural empowerment interventions and high levels of personal LGO are associated with high levels of feelings of impact, significance, self-determination, and competence in public employees. In contrast, low levels of LGO have a negative effect on this link, as structural empowerment is associated with lower levels of psychological empowerment. This supports the hypothesis that the individual orientation to learn and develop serves to make structural empowerment practices effective for generating psychological empowerment feelings in employees. It helps to complement and better understand the relationship between two types of empowerment previously studied without taking into account moderating effects (e.g., Cho & Faerman, 2010). By contrast, we found no significant moderation effects of PPGO or APGO in the main structuralpsychological empowerment relationship. These results invite some speculation. It is possible that structural empowerment is not reinforced by performance GOs because of its specific components. In this study, structural empowerment comprises training and knowledge, information sharing, and power to make decisions, and rewards. As stated by Fernandez and Moldogaziev (2013), empowerment practices have diverging effects. Perhaps training and knowledge, information sharing, and power to make decisions are not sufficient for individuals with high performance GO (PPGO or APGO) to be able to demonstrate their abilities, or they may not represent situations in which employees' capabilities can be displayed, well or badly, and assessed. Rather, these practices focus more on involving, motivating, and developing employees' skills. Thus, if there are no specific protocols or processes with which to assess and reward employees (apart from the bonus they receive for achieving individual or group targets that are actually private, monetary recognitions), the interaction of their willingness to demonstrate their aptitudes, be positively judged, compete against others, or avoid undesirable evaluations with the structural empowerment interventions cannot strengthen the effects of these empowerment practices. The structural-psychological empowerment link, therefore, remains insignificant regardless of employees' PPGO and APGO levels. Theoretically, it helps to shed more light on the moderating power of public employees' individual preferences for the structural-psychological empowerment relationship.
In addition to the theoretical contributions, this study also has some practical implications. First, the findings demonstrate that structural empowerment practices are not associated with employees' psychological empowerment. Therefore, in this context, merely putting structural empowerment practices in place is not sufficient to generate psychological empowerment in workers. This suggests that selecting employees who are able to promote their own psychological empowerment feelings is one possible way to cultivate a staff of psychologically empowered employees. Managers could also encourage this by offering employees the opportunity to shape their job according to their preferences and capabilities (Wrzesniewski & Dutton, 2001), as this job crafting procedure leads to a more meaningful view of work (Tims & Bakker, 2010;Wrzesniewski & Dutton, 2001), and may ultimately help psychological empowerment feelings, regardless of management interventions.
Second, the results suggest the value of employees' LGO in reinforcing the power of structural empowerment interventions to generate more psychologically empowered employees. Individuals who report higher levels of these empowerment feelings in the long term are likely to perform more effectively and behave better in their organizations (Jong & Ford, 2016). Thus, taking into account the NPM paradigm and its new ideas about the role of human resources and their involvement in and importance to improving services for citizens, finding a way to have more learning goal-oriented employees could help improve service delivery. In this respect, staff selection systems should be modified to take into account employees' competence and attitudes, such as LGO, as far as legal regulations on public employee selection allow it. Public employees generally have permanent contracts with the organization (Barba & Serrano, 2015). For instance, the Spanish public administration uses several different staff selection systems, but all of them involve written tests to prove the candidate has the knowledge and abilities required for the job. Psychological tests and interviews may be included, but they are rarely used. It would, therefore, be appropriate to routinely carry out tests to assess candidates' LGO, and thus match them with the knowledge and skills required for the job.

Limitations and Future Directions
The study has certain limitations, and also raises issues to address in future research. First, the data were obtained at a single point of time, so causality among the constructs cannot be assumed. A longitudinal study design could usefully assess the causal links. Another limitation is the use of self-reported and perception-based assessments. Most of the data were obtained through such sources, which are more likely to produce biased responses because of the social desirability and common method variance phenomena. Although we took data from two different sources and guaranteed the confidentiality of the responses, managers and employees may have over-reported desirable local government and individual characteristics, respectively. Future research should strive to mitigate this bias by introducing objective information.

Conclusion
In an attempt to go deeper into the structural and psychological empowerment link in the public sector, and the role of employees' GOs in this relationship, we provide empirical evidence grounded in the framework of local government. This study extends research in the field of empowerment by analyzing its dual perspective, finding an absence of any significant link between the two types, which suggests the importance of the intrinsic power of employees to generate their own feelings of empowerment. Furthermore, this study provides new empirical evidence by showing the key role LGO plays in making structural empowerment practices effective to create employees' feelings of psychological empowerment. To the best of our knowledge, our research is the first to expand understanding of the link between structural and psychological empowerment by analyzing the moderator role of GOs, incorporating motivation issues in this link. In sum, the findings suggest that the combination of structural empowerment practices and employees' high levels of LGO are important factors for creating psychologically empowered staff in public organizations.
11. I have a great deal of control over what happens in my department. 12. I have significant influence over what happens in my department.

Goal Orientation (Employee Survey)
Please fill in the number that represents how you feel about these questions (1 = strongly disagree, 7 = strongly agree).
LGO 1. I am willing to select a challenging work assignment that I can learn a lot from. 2. I often look for opportunities to develop new skills and knowledge. 3. I enjoy challenging and difficult tasks at work where I'll learn new skills. 4. For me, development of my work ability is important enough to take risks. 5. I prefer to work in situations that require a high level of ability and talent. PPGO 6. I like to show that I can perform better than my coworkers. 7. I try to figure out what it takes to prove my ability to others at work 8. I enjoy it when others at work are aware of how well I am doing. 9. I prefer to work on projects where I can prove my ability to others. APGO 10. I would avoid taking on a new task of there was a chance that I would appear rather incompetent to others. 11. Avoiding a show of low ability is more important to me than learning a new skill. 12. I'm concerned about taking on a task at work if my performance would reveal that I had low ability. 13. I prefer to avoid situations at work where I might perform poorly.

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) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: The authors appreciate the financial support for this research from the Ministerio de Economía y Competitividad (Ref. ECO2015-66671-P MINECO/ FEDER) and from the Universitat Jaume I (Ref. UJI-B2017-21) (SPAIN).