If you build fences around employees, you get sheep
Do empowering leadership, team reflexivity and accountability influence personal initiativeand how do these relate?
MSc Business Administration
Specialization Leadership and Management Master Thesis Final
Author: Milou Heijmerink, 10647171 Primary Supervisor: Renske E. van Geffen Secondary Supervisor: Hannah A. Berkers
June 24th 2016
Statement of Originality
This document is written by Milou Heijmerink, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
___________________________________________________________________________
Abstract
Due to the rapidly changing environment it is important that employees engage in personal initiative, which strongly contributes to the survival of the organization. Therefore the purpose of this research is to examine situational antecedents, which organizations and leaders could manipulate in order to enhance employee personal initiative. A model is proposed and tested with PROCESS analysis, for which data was collected from multiple Dutch teams with a leader (N = 250). In line with previous work the research established positive effects of empowering leadership and team reflexivity on personal initiative. In addition it was found that employee felt accountability partially mediates the relationship between empowering leadership and personal initiative and team reflexivity and personal initiative. The research also expected to find a curvilinear relationship between accountability and personal initiative, however the findings did not support this notion. The findings indicate that organizations and leaders, who focus on empowering leadership and team reflexivity, will positively influence employee personal initiative. The theoretical contributions are highlighted and the research concludes with acknowledging the research limitations and providing fruitful directions for future research and application.
Keywords: empowering leadership; team reflexivity; personal initiative; perceived accountability.
Table of Contents
INDEX OF FIGURES AND TABLES ... 5
INDEX OF APPENDICES ... 5
1 INTRODUCTION ... 6
2 THEORETICAL BACKGROUND ... 10
2.1. Personal Initiative and the influence of Empowering Leadership ... 10
2.2 Empowering Leadership and Accountability ... 14
2.2.1 Team Reflexivity and Personal Initiative ... 17
2.2.5 Team Reflexivity and Accountability ... 19
2.2.3 Accountability and Personal Initiative ... 22
3 METHODS ... 29 3.1 RESEARCH DESIGN ... 29 3.2 MEASURES ... 31 3.3 DATA ANALYSIS ... 32 3.4 PREDICTIONS ... 33 4 RESULTS ... 34 4.1 DESCRIPTIVE STATISTICS ... 34 4.2 NON-‐LINEAR RELATIONSHIP ... 35 4.3 MEDIATION ... 36 4.3.1 Empowering Leadership ... 37 4.3.2 Team Reflexivity ... 38
5 DISCUSSION AND CONCLUSION ... 39
5.1 DISCUSSION ... 39
5.2 Research Contributions ... 42
5.2 PRACTICAL IMPLICATIONS ... 44
5.3 LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH ... 45
5.4 CONCLUSION ... 51
REFERENCES ... 53
APPENDIX A QUESTIONNAIRE INSTRUCTIONS ... 62
Index of Figures and Tables Figure 1. The research model
Table 4.1. Descriptive Statistics and Correlations
Table 4.2. Results of PROCESS analysis with independent variable Empowering Leadership Table 4.2. Results of PROCESS analysis with independent variable Team Reflexivity
Index of Appendices
Appendix A Questionnaire Instructions Appendix B Questionnaire Items
1 Introduction
The title of this research, ‘If you build fences around employees, you get sheep’ is related to a quote of William L. McKnight, who strongly believed in delegation of responsibility and employee ownership of initiatives (Govindarajan & Srinivas, 2013). The former CEO of 3M Company - from 1949 to 1966 - transformed the organization into one of the world’s most innovative organizations by creating a culture that encourages employee initiative and innovation through tolerating mistakes and encouraging success. McKnight’s ideology that success could only be achieved through people taking initiative induced him to provide employees with opportunities, support and freedom to think and act(3M Company, 2002)Specifically, 3M creates a wide variety of systems, structures and work environments. For instance the ‘15% rule’, which means employees can spend up to 15% of their time pursuing projects of their own choice and to look for unscripted opportunities (Govindarajan & Srinivas, 2013). Another example is the Technical Council, where scientists periodically share knowledge and build relationships, while discussing progress on their projects. This is the forum where the Post-It Note was born (Govindarajan & Srinivas, 2013). McKnight established fundamental operating principles for 3M including belief in people, trust in people and willingness to let people try new things (The 3M Story, 2002). This led to the creation of an outstanding worldwide organization, driven by proactive employees exercising their personal initiative and resourcefulness to overcome obstacles, because they get provided the room they need (Gopinath, 2013).
Nowadays, the fast-changing, uncertain business environment demands for organizations such as 3M Company, which are able to proactively stay ahead of the global competitive pressures (Dewettinck & van Ameijde, 2011)Consequently, organizational structures and job designs have become more flexible than ever before, implying the need for employee proactive behavior without the need for constant close supervision (Grant &
Parker, 2009)Moreover, in the current knowledge economies with complex work tasks, organizations greatly rely upon employee competence and proactivity in order to boost creativity, innovation and change (Grant & Ashford, 2008)These circumstances raise the question how to get flexible and responsible employees who go beyond their narrow task requirements and who approach work proactively (Frese, 2008).
Proactive behavior indicates anticipated actions of employees to impact themselves, their environment or both (Grant & Ashford, 2008). There are different ways in which employees enact proactive behaviors, for instance expressing voice (LePine & Van Dyne, 1998), job crafting (Wrzesniewski & Dutton, 2001) or issue selling (Dutton, Ashford, O'Neill, & Lawrence, 2001)These future-focused, mindful, change-oriented ways of behaving are important because they can prevent potential problems and enable the exploitation of opportunities for future development, thereby resulting in greater individual or organizational effectiveness (Grant & Ashford, 2008; Wu & Parker, 2011).
To gain a deeper understanding of how to cultivate employee proactivity, this research will elaborate on a specific proactive behavior, namely personal initiative (Frese & Fay, 2001). Behavior is context specific and therefore any employee can engage in personal initiative if the situation allows for it (Bolino, Valcea, & Harvey, 2010). Therefore relevant contextual antecedents of personal initiative will be considered including the underlying psychological mechanisms that foster the intended outcome of employee personal initiative.
Contextual factors that affect the development of personal initiative are for example control at work (Menon, 2001), support from the leader (Ahearne, Mathieu, & Rapp, 2005) and accountability (Parker, Williams, & Turner, 2006). Additionally, mediating mechanisms that have been examined are for instance self-efficacy (Parker et al., 2006) and commitment or affect at work (Den Hartog & Belschak, 2007). Research has highlighted the importance of leadership for multiple employee outcomes (e.g. Den Hartog & Koopman, 2001). This
research proposes the exceptional power of an empowering leader, who is able to establish most of the previously mentioned antecedents of personal initiative and therefore indicating a positive effect on employee personal initiative (Seibert, Silver, & Randolph, 2004).
While personal initiative often contributes to both employee and organizational performance, it is suggested that employees make a trade-off when deciding to engage in personal initiative in terms of potential costs and benefits for their image (Grant & Ashford, 2008). A proposed influencer on this evaluation is the perceived level of accountability(Hall et al., 2006)Accountability has even been considered a precondition for proactive behavior and organizational effectiveness, as it guides employees in a prescribed manner (Lerner & Tetlock, 1999). The need for accountability in the work environment is underscored by corporate scandals, as the Enron and its accounting firm scandal or the Volkswagen emissions scandal. Moreover, it is expected that employees are increasingly held accountable for work activities and outcomes, which intensifies the need for more research regarding accountability mechanisms in actual work context (Hall et al., 2006).
Furthermore, the current dynamic and competitive business environment contributes to other organizational changes (Den Hartog & Koopman, 2001). Among these changes is an increased importance of teams and self-managed teams (Bolino et al., 2010). This implies a transfer of the leadership responsibilities to the team or the employee to encourage personal initiative (Manz & Sims, 1980). The present research will focus on the interactive process of team reflexivity, where tasks and outcomes are discussed and modified to improve team effectiveness (Schippers, Den Hartog, & Koopman, 2007). Since team reflexivity is an active process of reflection, planning and adaptation, there is reason to expect that reflexivity has an influence on employee personal initiative, which is mindful anticipated, planned and action-oriented behavior (De Dreu, 2002; Fay & Frese, 2001).
research question arises: Do empowering leadership, team reflexivity and accountability influence personal initiative and how do these relate? To answer this question, data from multiple teams with a leader will be analyzed, which is going to be collected through questionnaires. Understanding the situational and motivational factors that underpin employee personal initiative enables organizations and leaders to better stimulate this increasingly critical behavior. The research model and indications of the proposed hypotheses are depicted in Figure. 1.
Figure 1. The research model
Additionally, it adds to the current knowledge about accountability in organizational context and assumes it as fruitful mechanism for directing the employee in ‘pro-company’ personal initiative, through felt accountability (Bolino et al., 2010). Furthermore, the present research contributes to the research field by adopting a thorough conceptualization of empowering leadership, based on delegating authority and support (Conger & Kanungo, 1988). Next, this research contributes to the rather limited research concerning team reflexivity. Lastly, steps are taken towards answering the call of (Grant & Ashford, 2008) for future attention to the relationship between situational antecedents of personal initiative, respectively accountability and autonomy.
The following research begins with the theoretical background where employee personal initiative is defined. The relationship with the different situational or motivational factors is provided, each ending with a hypothesis. The subsequent section describes the research methodology. In the result section the data will be analyzed and reported, followed by interpretations and elaborations in the discussion. Additionally, the discussion considers the research contributions and limitations, as well as directions for future research. Finally, conclusions are drawn concerning the effect of empowering leadership, team reflexivity and accountability on employee personal initiative.
2 Theoretical Background
The ensuing section will start with the development and description of employee personal initiative. It will be followed with explanations for the proposed relationships concerning empowering leadership and thereafter with team reflexivity. In addition will be specified on the linear and non-linear relationship between accountability and personal initiative. Each section ends with a hypothesis, which helps to establish the answer to the main research question.
2.1. Personal Initiative and the influence of Empowering Leadership
Organizational research regarding proactivity has been introduced with the concept of proactive personality, defining it as a stable desire of the employee to bring about change (Grant & Ashford, 2008). However, that does not inform which behaviors should be classified as proactive (Crant, 2000). Also, while proactive personality refers to the employees’ dispositional characteristics regarding the intention to engage in proactive behavior, it may not be assumed that proactive behavior will occur (Grant & Ashford, 2008). Consequently, Frese and Fay (2001) constructed personal initiative as proactive work
behavior. The aim of such behaviors is to improve given work methods and procedures as well as developing personal conditions for meeting future work demands (Frese, Fay, Hilburger, Leng, & Tag, 1997). The focus on personal initiative and situational antecedents enables to understand under which circumstances, both employees with and without a proactive personality, may be motivated to engage in personal initiative.
Before the study elaborates on the factors that enhance employee personal initiative, it is important to specify what employee personal initiative entails. Personal initiative is a goal-oriented behavior defined by three key elements, which reinforce each other (Fay & Sonnentag, 2002). First, it is self-starting behavior, implying the pursuit of self-set goals in contrast to assigned goals (Frese & Fay, 2001). Second, personal initiative is future-focused. It is proactive through anticipation of future opportunities or challenges, rather than waiting for demand to become too high. Third, it is change-oriented, which often means it includes encountering difficulties (Crant, 2000). To overcome these obstacles, personal initiative is also defined as being persistent (Frese & Fay, 2001). Thus, personal initiative is self-starting, proactive and persistent behavior, which could mean spending additional time and energy in the work job or going beyond the prescribed contents of the job (Rank, Pace, & Frese, 2004). The fundamental elements of personal initiative and the crucial importance of this behavior for organizational successfulness, indicate that the organization or the leader should not be the one to direct and decide (Mathieu, Maynard, Rapp, & Gilson, 2008). Moreover, since the employee requires some control over the work job to engage in personal initiative, the role of the leader is to provide employees with considerable freedom concerning how to initiate and carry out tasks (Frese & Fay, 2001; Zhang & Bartol, 2010). In addition, the leader has to provide an environment through which employees are encouraged to anticipate, plan and act to address the economic demands and opportunities, hence show personal initiative (Grant & Parker, 2009).
There are multiple reasons to expect that empowering leadership has a positive influence on personal initiative, because empowering leadership involves sharing power and responsibility with a view toward enhancing employees’ self-motivation and investment in their work (Kirkman & Rosen, 1999). This is done by highlighting the significance of the employee’s job, expressing confidence in the employee’s capabilities, providing greater decision-making autonomy, and removing hindrances to performance (Ahearne et al., 2005). This shows that empowering leadership differs from the traditional leadership practices in organizations, which are more hierarchical (Dewettinck & van Ameijde, 2011). In addition, it shows the potential of empowering leadership in reshaping the structure of the organization to become more in line with the environmental demands and increase the chances of survival (Grant & Parker, 2009). Because employees are enabled to make decisions and implement actions without the direct supervision of the leader, employees are able to become more adaptive and receptive to the dynamic environment (Spreitzer, 1995). Then, employees are able to manage the situation in which they find themselves by engaging in personal initiative (Crant, 2000). Thus employees are more likely to act in advance to prevent problems, introduce greater structure, and improve their tasks (Grant & Ashford, 2008).
Empowering leadership is highly capable of enhancing employee personal initiative, because it is mainly aimed at employee development (Pearce et al., 2003). The goal is to shape employees into effective self-leaders that are capable of initiative, creativity, and acting on their own. Therefore, the empowering leader enacts in behaviors such as coaching, informing, leading by example, providing support and concern and participative decision-making (Dewettinck & van Ameijde, 2011). Together these behaviors establish an environment of trust and safety, which result in more positive employee work-attitudes (Kark, Shamir, & Chen, 2003). Moreover, in a safe environment employees are not afraid to engage in interpersonal risks involved in learning behaviors such as discussing errors or
displaying personal initiative (Edmondson, 1999).
The influence of empowering leadership on employee decision-making and behavior can be explained via engagement in the social learning process (Bandura, 1988). In states that employees observe, interact, and summarize norms on the basis of leader behavior that develops into a common understanding of that behavior (Wallace, Johnson, Mathe, & Paul, 2011). The empowering leader models certain behaviors, which help acquiring and developing new employee behaviors (Bandura, 1988). For instance, empowering leadership promotes and encourages opportunity thinking, self-leadership, self-rewards, and participative goal setting, which are adopted by the employee (Pearce et al., 2003). Therefore, given the self-set nature of personal initiative (Frese & Fay, 2001), it is likely than an empowering leader establishes the right example and environment for an employee to engage in personal initiative.
Another important mechanism of empowering leadership that influences employee behavior is through the use of goals (Pearce et al., 2003). Goals have a strong motivational effect, because they provide a sense of purpose and direction (Bandura, 1988). Empowering leadership is focused on participative goal setting, and the focus on self-leadership may be a clear added value, as it motivates the employee to set appropriate goals (Pearce et al., 2003). In addition, it is proposed that empowering leadership practices enhance employee’s internalization of proactive expectations and external proactivity goals (Parker et al., 2010). This is the extent to which individuals see it as important and ‘their job’ to be proactive and in return it is expected that employees take responsibility for the proactive goals (Wu & Parker, 2011).
The underlying mechanism responsible for the effect on employee behavior is based on employee self-efficacy (Bandura, 1988). The empowering leadership practices enhance the employee’s perception of self-efficacy, for example through providing autonomy or
expressing belief in employee competence (Zhang & Bartol, 2010). Self-efficacy is the employee’s belief in his or her capability to successfully perform tasks (Bandura, 1988). It affects employee behavior, since employees tend to avoid behaviors that are believed to exceed their capability (Menon, 2001). Thus, a strong belief in own capabilities has a positive effect on the choices employees make, the effort they put forth, their perseverance in the face of difficulties and setbacks and resilience after encountering setbacks (Bandura, 1988). For instance, employees with a high belief of self-efficacy approach difficult tasks not as threats to be avoided, but rather as challenges to be mastered. Additionally, high self-efficacy fosters interest and strong involvement in activities (Bandura, 1988). In contrast, when self-efficacy is low, employees are less likely to persist at tasks that require significant effort (Kark et al., 2003). This indicates that empowering leadership is able to enhance motivation of the employee for setting proactive goals, to enhance investment and persistence in reaching those goals, which reflects the foundation of personal initiative (Gagné & Deci, 2005).
In summary, an empowering leader provides employee autonomy, establishes a safe environment, models for personal initiative, enhances employee proactive goal setting and goal attainment and overall enhances employee self-efficacy. It can be concluded that an empowering leader signals to employees that they have both the opportunity and ability to actively shape and create the work they perform by engaging in personal initiative (Grant & Ashford, 2008). This resumes in the following hypothesis.
H1: Empowering leadership is positively related to employee personal initiative.
2.2 Empowering Leadership and Accountability
Accountability is perhaps one of the most fundamental organizational constructs and essential for effective running of organizations (Hall & Ferris, 2011). Absence of the accountability construct would lead to chaos and disruption of organizations, as employees would have the
freedom to do whatever they desired, with no controls or sanctions for their behavior (Hall, Zinko, Perryman, & Ferris, 2009). Therefore the accountability mechanism is based on the effort of organizations to shape employee behavior in the organizationally prescribed direction (Hall et al., 2009).
There exist multiple definitions of accountability, but overall three aspects are predominant, namely a social relationship, and evaluation of others, and the potential for rewards or sanctions (Frink & Klimoski, 2004). Accountability in organizations refers to the situation in which some audiences, including one’s self, hold others answerable for their decisions and behaviors (Ferris et al., 1997; Hall et al., 2009). Also, there are certain standards or expectations against which the employee’s behaviors are compared (Frink & Klimoski, 2004). It implies the belief of the employee to justify behaviors, and in return to expect potential rewards or sanctions. These rewards or sanctions can be explicit or implicit, for instance social normative expectations, and also objective or subjective (Frink & Klimoski, 2004).
Looking at the two different constructs – first, empowering leadership providing employee freedom, and second, accountability establishing employee constrain – they seem to be working in the opposite direction. While empowering leadership is associated with less constrain for the employee, it could be seen as an “enabling” factor, because it enables own behavior and initiative (Wu & Parker, 2011; Conger & Kanungo, 1988). Accountability, on the other hand, can be seen as an “expected to” factor, because it involves stronger situational direction for behavior (Wu & Parker, 2011). Nonetheless, the constructs are not that dichotomous when is acknowledged that within all accountability relationships it is a necessary precondition to have some autonomy or control on decisions and behavior (Lindkvist & Llewellyn, 2003). Namely, when the means for certain goals are not provided, the employee has the freedom to construct the means and hence, can be hold accountable for
it (Seibert et al., 2004). After all, that is precisely the reason why organizations put accountability mechanisms in place (Frink & Klimoski, 2004). Put differently, without autonomy, one simply cannot experience accountability.
This means that when the empowering leader provides the employee with authority concerning the work job, the responsibilities for that work job are also delegated, establishing accountability (Seibert et al., 2004). Then, in order for accountability to have the intended influence on employee behavior, the employee has to feel accountable (Frink & Klimoski, 2004). For example, in a situation where autonomy would be high, but employee felt accountability would be low, it is not clear what is being expected from the employee, what the employee provided autonomy is about and what the evaluation audiences or methods are (Wallace et al., 2011). As a consequence, employees who do not feel accountable may take advantage of their autonomy and empowerment resulting in potential adverse outcomes (Wallace et al., 2011). The preceding indicates that solely providing autonomy and responsibility not necessarily leads to employee felt accountability (Frink & Klimoski, 2004). For the employee to feel accountable the employee needs to know the standards or expectations of desired behavior and the employee needs to expect potential rewards or sanctions, which makes it meaningful for the employee to comply with the internal pressures (Ferris et al., 1997; Hall et al., 2006). Based on social learning theory (Bandura, 1988), it is recognized that the values of the leader are reflected downward as vicarious learning, providing the expectation for desired behavior. It informs employees what behaviors are likely to be rewarded or sanctioned (Frink & Klimoski, 2004). When the empowering leader recognizes compliance of the employee to the desired behavior, this might result in higher supervisor ratings, increased job satisfaction or promotion (Hall et al., 2009). Additionally, Frink and Klimoski (2004) suggest that the soft influence tactics may produce a sense of obligation, which in turn insures accountability.
To conclude, because empowering leaders utilize softer influence tactics and motivate employees towards self-actualization, employees feel accountable for their work job (Bommer, Rich, & Rubin, 2005). In this sense, employees are motivated, committed and satisfied and assist the organization in achieving its objectives. In other words, the structure – authority and responsibility – in combination with the actions – encourage self-leadership – of an empowering leader, come hand in hand with increased felt accountability for the employee.
H2: Empowering leadership is positively related to employee accountability
2.2.1 Team Reflexivity and Personal Initiative
An important principle identified for effective team working and innovative action is the team process of team reflexivity (De Dreu, 2007). Team reflexivity is defined as “the extend to which group members overtly reflect upon, and communicate about the group’s objectives, strategies (e.g. decision-making) and processes (e.g. communication), and adapt them to current or anticipated circumstances” (West, 2000, p.3, in Schippers et al., 2008). Thus, when teams reflect on and modify their functioning, teams will be more effective. In the work environment, reflexivity is an important factor for team and organizational sense making and is central to learning from experience (Schippers, Den Hartog, Koopman, & van Knippenberg, 2008). Also, research found a positive relation between team reflexivity and team satisfaction, commitment and performance (Schippers, Den Hartog, Koopman, & Wienk, 2003). It is noted that reflexive teams plan more in detail, pay more attention to long-term outcomes and are better able to respond to the environment (Widmer et al., 2009). Moreover, teams that are not reflexive are prone to be reactive rather than proactive, indicating the potential positive effect of team reflexivity on employee proactivity.
Team reflexivity can be seen as a relational activity involving necessary interaction and observable behaviors, while sharing and discussing views and ideas among employees
(Schippers et al., 2008). Following social learning theory (Bandura, 1988), the employee will process these observations and perceived information and then model their behavior accordingly. This means the team reflexivity teaches employees to reflect, plan and adapt, which is integrated into their behavior (Wallace et al., 2011). Additionally, reflexive teams are committed to continuous improvements (Schippers, Den Hartog & Koopman, 2007), which would have a positive effect on employee personal initiative, which is likewise focused on improvement (Crant, 2000). Thus, the expectations for behavior, as developed and enforced during team reflexivity, are likely to have a great impact on employee behavior (Frink & Klimoski, 2004).
To continue with the effect of team reflexivity on employee personal initiative, it is important to know that knowledge concerning how to carry out personal initiative, is likely to be learned through observation and experience since there are no clear rules or prescriptions for engaging in personal initiative (Grant & Ashford, 2008). Therefore this research turns to the observable behaviors in the reflexivity process. The observable behaviors under team reflexivity are based in the iterative collective process of reflection, planning and action (Widmer et al., 2009). Comparable to this process is the proactivity process, which is based on the sequence and interrelated phases of anticipation, planning and action (Grant & Ashford, 2008). In other words, both reflection and proactivity processes identify and solve problems and are action-focused (Crant, 2000). Reflection is the consideration of work-related issues. In order to establish changes, adaption needs to occur and planning could be seen as the bridge between first reflection and second adaptation (Widmer et al., 2009). Just as the process for proactivity, the planning phase contains establishing goals and ways to achieve those goals (Grant & Ashford, 2008; Widmer et al., 2009). The action phase in both processes relates to the goal-directed behaviors for achieving the desired changes, hence directed toward future impact. Thus, the observed behaviors intended for enhancing team
effectiveness, teach the employee to reflect, plan and adapt own behavior, which implies and is likely to result in the demonstration of personal initiative.
In addition, team reflexivity makes the employee aware of the environmental challenges and how to overcome them. Specifically, employees reflect on the appropriateness of their objectives and strategies, the efficiency of their processes and the reality of their changing environment (Widmer et al., 2009). Knowledge about these circumstances is important for employee personal initiative, which involves challenging the status quo rather than passively adapting to present conditions (Crant, 2000). When employees direct their actions toward future impact, they are mindful of the effects of their actions on themselves and their environments (Grant & Ashford, 2008). In addition, not only the short-term impact of their actions is considered, but often also the potential long-term impact of their actions. They are acting with anticipation to prevent future problems and seize future opportunities (Frese & Fay, 2001).
In conclusion, the observable behaviors of fellow employees focused on reflection, planning and action, set the expectation for appropriate behavior (Bandura, 1988). Employees interpret this and integrate it into their own behavior (Wallace et al., 2011). Next to that, because the employee becomes aware of the environment, the employee knows what the challenges are and how to overcome them and becomes mindful (Grant & Ashford, 2008). This means that reflection, planning and action within the team based on past actions, helps the individual employee to anticipate, plan and act towards future impact.
H3: Team reflexivity is positively related to personal initiative.
2.2.5 Team Reflexivity and Accountability
As indicated before, team reflexivity is a highly interactive activity that involves sharing and discussing views and ideas among coworkers (Schippers et al., 2008). When team reflexivity occurs and employees interact with each other on a regular basis, certain patterns develop
(Frink & Klimoski, 2004). It sets expectations about what someone is supposed to do under a particular set of circumstances (McGrath, 1984 in Frink & Klimoski, 2004). In effect, these expectations are the rules for appropriate behavior. So, it is likely that the interaction processes between employees during team reflexivity will set clear expectations for employee behavior (Frink & Klimoski, 2004). These perceptions related to “how things are done around here”, influence the behavior of the employee (Hall & Ferris, 2011). These are the group norms and from an informal accountability mechanism (Frink & Klimoski, 2004). So, accountability is a perceptual condition, based on objective factors and based on a cognitive factor concerning what employees think that they need to do (Wallace et al., 2011). In the organizational context informal accountability is subject to interpersonal relationships, which are developed over time shaping the nature and extent of the perceived accountability (Frink & Klimoski, 2004). Moreover, Hall and Ferris (2011) suggest that the informal norms and expectations might have stronger influence on employee behavior, than the formal rules of the workplace.
Earlier it was stated that accountability stimulates organizational order by creating awareness of task responsibilities, of acceptable methods for task achievements and helps to identify how behavior will be evaluated. Hence, through team reflexivity, the employee acquires awareness of task responsibilities and of acceptable methods for task achievements. In addition, because of the collective interaction component of team reflexivity, employees are able to identify and compare rewards and sanctions for behavior (Parker, 2014). Through interaction and individual action employees are observed, where performance discrepancies will be visible, and where feedback and learning can be provided, implying rewards (De Jong & Elfring, 2010). At the same time accountability serves as guidance, but also as a certain boundary condition for employee behavior, in line with team expectations (Hall et al., 2006). Accordingly, the first association between team reflexivity and employee accountability is
the provision of clear expectations about behavior and possible rewards, which promotes accountability (Hal et al., 2003).
The second association is due to the goal of team reflexivity. Reflexive teams are committed to continuous improvements and increasing team innovation (De Dreu, 2002). During team reflexivity, teams will often evaluate if the employees are working efficiently in order to benefit future activities of the team (Schippers et al., 2007). Employees expect others to elucidate their actions, in order to enhance future performance of the team (De Jong & Elfring, 2010). This is possible, because reflexivity is not only about reflection and planning, but also about actions, which make it visible for others (Widmer et al., 2009). The actions carried out by the team or its members in return can lead to further reflexivity, which implies the iterative and ongoing process (West, 2000, in Widmer et al., 2009). This means that employees have both the reason and opportunity to observe and evaluate the other employee’s actions (Frink & Klimoski, 2004). So, employees closely scrutinize the efforts of other coworkers with the goal to enhance team effectiveness, establishing accountability (W. A. Hochwarter, Perrewé, Hall, & Ferris, 2005).
Previous research concerning accountability often studied how subjects participate in an experiment in which one group is held accountable and the other one is not (Frink & Klimoski, 2004; Lerner & Tetlock, 1999). However, within a team context, employees can also hold each other accountable (van Mierlo, Rutte, Vermunt, Kompier, & Doorewaard, 2006). Given that team reflexivity consists of attention, awareness, monitoring, and evaluation of the object of reflection, team reflexivity is a higher-level construct that is likely to have an influence on employees’ cognitions (Shin, 2014). More specifically, employees collectively control their own behaviors via a variety of mechanisms, one of which the social structure of the team (Van Mierlo et al., 2006). Hence, teams often develop their own system of norms and rules and subsequently enforce these team-based ‘control systems’ on each
other (Van Mierlo et al., 2006). Then, employees are aware of the expectations and evaluations for behavior, because employees work under the same social control system, which is fueled through every interaction during team reflexivity.
In conclusion, interaction ensures employees to become mutually aware of the norms, expectations or rules of behavior in particular circumstances (Frink & Klimoski, 2004). When the employees together reflect on past behavior, this will establish certain behavioral standards and expectations; moreover it serves as a “behavioral check on the appropriate application of effort and resources” (Wallace et al., 2011, p. 843). Second, because the process of reflexivity is intended for team effectiveness, employees will scrutinize each other’s behavior, increasing accountability. Third, employees are mutually aware of the accountability mechanisms, its evaluative audiences and the opportunities for rewards and sanctions. This results into the following hypothesis.
H4: Team reflexivity is positively related to accountability.
2.2.3 Accountability and Personal Initiative
As acknowledged before, accountability mechanisms could be seen as a control mechanism from the organization to lead employee behavior in prescribed directions (Hall et al., 2009). For the employee, accountability refers to the expectation of being a subject of evaluation and is based on the pressure to defend one’s actions to others (Royle, Hall, Hochwarter, Perrewé, & Ferris, 2005). Additionally, research has found that particular forms of accountability could motivate more effortful and self-critical thinking (Grant & Parker, 2009). This is
because employees construct their decisions to correspond to the outcomes they desire from
the evaluating audience (Tetlock, 1985 in Royle et al., 2005). Research with a traditional focus indicates that in response to the expected evaluation, an employee will set goals to enhance performance. A way to do so is by doing more than one is being expected to do, thus
engaging in personal initiative (Crant, 2000). Other research has also indicated that the goals set to weather the next moment of evaluation, could serve as an alternative function, namely for impression management (Frink & Ferris, 1998).
Before expressing personal initiative, employees will consider the potential costs and benefits of that proactive behavior for their image, job performance, career development, and other relevant outcomes (Crant, 2000). Outcomes of primary importance are the ones concerning the social costs or benefits; therefore the employee focus is on self-presentation or impression management (Crant, 2000; Royle et al., 2005). It has been suggested that accountability, and the evaluative dimension of others and one’s self, has implications for the employee’s private self-image (Frink & Ferris, 1998). This implies that the potential image costs or benefits due personal initiative are considered, which is influenced by the level of perceived accountability (Royle et al., 2005).
As stated before, personal initiative is self-starting future-oriented behavior comprising of doing more than is expected and, what the employee appears to be personally responsible for (Frese & Fay, 2001). Therefore, when the employee expresses personal initiative that fails, under-performs or makes an error, it will be difficult to blame the outcomes on external circumstances, hence pointing the spotlight on the employee (Grant & Ashford, 2008). In return, since others observe and interpret employee’s behavior as distinctive for the employee’s values and beliefs, the employee’s image will be influenced (Lerner & Tetlock, 1999; Grant & Ashford, 2008). Accordingly, when employees are not held accountable, the rational mechanism weighing costs against benefits for their image, will promote especially higher image costs (Frink & Ferris, 1998). Hence, proactive behaviors such as personal initiative, is safest to be avoided under the circumstances of low accountability, by avoiding attention in order to protect one’s image (Grant & Ashford, 2008).
In contrast, when people do get held accountable, impression management theory suggests a contrasting reaction (Hui, Lam, & Law, 2000). The mechanism is based on the same cost and benefit analysis. However, under high accountability the perceived image benefits will be higher than the perceived image costs of personal initiative (Grant & Ashford, 2008). Specifically, because the employee will be held personally responsible for its actions regardless of whether he or she behaves reactively or proactively, the employee faces little additional risk in employing personal initiative (Grant & Ashford, 2008). Because employees are concerned about other’s impressions in the evaluative situation, they are motivated to engage in anticipatory proactive action to create favorable impressions (Grant & Ashford, 2008). Given the already projected attention, employees may as well work in advance, bringing attention to the contributions that they make to the organization (Hall & Ferris, 2011). So, accountability motivates employees to engage in personal initiative by reducing the perceived image costs of personal initiative and providing the opportunity to improve one’s performance or image (Bolino et al., 2010; Grant & Parker, 2009).
Despite the positive notion of personal initiative of the employee, it has been acknowledged that personal initiative could concurrently harm other employees or the organization, when actions are intended to benefit personal goals (Griffin & Lopez, 2005). Considering that “employees often engage in self-starting, anticipatory actions to benefit only themselves or have a destructive rather than constructive effect on their organizations, personal initiative does not paint a complete picture of the nature of proactive behavior” (Grant & Ashford, 2008, p.8). However, when the employee is held accountable one must not forget the dependence on the evaluation of others and the potential for receiving rewards or sanctions (Frink & Klimoski, 2004). Under accountability, the employee will know what is being expected from him or her, moreover the employee will know which behavior will be rewarded or sanctioned, and is therefore motivated to adapt to that (Wallace et al., 2011).
This shows how accountability has the capability to direct the employee in the intended direction of personal initiative as ‘pro-company’ behavior (Frese & Fay, 2001; Lerner & Tetlock, 1999). Therefore this research states that there is a positive relationship between employee accountability and employee’s personal initiative, because the perceived attention and evaluation lower the costs, while increasing the potential benefits for anticipatory personal initiative to enhance employee’s performance or image.
H5a: Employee accountability is positively related to employee’s personal initiative.
The previous section has its reasoning based on a linear relationship between employee accountability and employee’s personal initiative. This is in line with the notion that more accountability is better, because it produces organizationally desired outcomes (Frink & Klimoski, 2004). Indeed, the traditional view suggests that outcomes related to accountability are generally positive (Hall et al., 2006). Namely, because accountability is predicted on the assumption that employees have to provide justification for their behavior and evidence of their contribution, employees will strive for means or behavior necessary to ensure that favorable outcomes result. Similarly, it is stated that accountability buffers employee stress, because accountability requires individuals to acquire control (Hall et al., 2006). Nevertheless, contradicting findings exist regarding the relationship between accountability and employee outcomes (Lerner & Tetlock, 1999; Frink & Klimoski, 2004). While Hall et al. (2009) found a positive linear relationship between accountability and self-reported organizational citizenship behavior, which is related to personal initiative (Grant & Ashford, 2008), Wallace et al. (2011) suggested that extreme levels of accountability might serve as a stressor and therefore hurt employee performance. For example, Hall et al. (2006) propose
that accountability could serve as a stressor leading to job tension, job satisfaction and emotional exhaustion.
Thus, while accountability is important for the effective operation of any organization, it is not solely positive and has a ‘dark side’ (Royle et al., 2005). Also, too much task accountability could be perceived as illegitimate and may even backfire beneficial effects (Lerner & Tetlock, 1999). Hence, employees need some accountability to be motivated and to stay on the task, but too much accountability will not will produce undesirable effects and could lead to decline in intrinsic motivation or even counterproductive work behaviors (Lerner & Tetlock, 1999). In addition, researchers have criticized the assumption of linearity in organizational processes, whereby potential significant findings are not discovered (Ferris et al., 2006 in Hall & Ferris, 2011). Based on the ‘dark side’ of accountability and linearity critics, a non-linear relationship is suggested.
Hall and Ferris (2011) found a non-linear (U-shaped) relationship between accountability and extra-role behavior. While these researchers indicated that extra-role behavior was high both with low accountability, due to having extra time, and with high accountability, to serve impression management; with moderate levels of accountability the contextual performance dropped. Contrastingly, this research explains reasoning for why results could be the exact opposite (inverted U-shaped), suggesting a significant relationship between accountability and employee personal initiative for moderate levels of accountability, but a non-significant relationship under low or high levels of accountability. This will be elaborated on subsequently.
The reasons for this non-linear relationship stem from a resource perspective, which proposes that employees seek to obtain and to protect their resources in that when resources are threatened, employees tend to experience stress (Hobfoll, 1989, in Bolino et al., 2010). Considering accountability mechanisms often arise from multiple entities with divergent
needs (Frink & Ferris, 1998) employees need to utilize integrative complexity of thoughts to secure positive evaluations from the multiple entities (Lerner & Tetlock, 1999). Especially since employees will try to uphold a good impression (Bolino et al., 2010). Next to that are employees’ multiple accountabilities often in conflict with each other, which requires employees to prioritize in decision-making, possibly inducing role overload (Hall et al., 2006). In conclusion, because high levels of accountability require employees to utilize more energy and interpersonal resources than low levels of accountability, high accountability likely facilitates less favorable individual and organizational outcomes (Hall et al., 2006).
Regarding the relation to personal initiative, one has to take back into account that personal initiative is self-starting proactive behavior (Frese & Fay, 2001), which often involves going beyond the employee’s everyday job responsibilities (Bolino et al., 2010). Namely, whether it is in-role or extra-role, proactive behavior entails anticipation, planning, and action directed towards future impact (Grant & Ashford, 2008). As such, personal initiative requires extra work or taking additional responsibilities (Bolino et al., 2010). Higher accountability demands the employee to thoughtfully distribute the finite resources (Hall et al., 2006), as well as diminishing the resources available for personal initiative.
Another reason for not engaging in personal initiative under high accountability relates to the employee expectations of potential rewards and sanctions, which in return influences their behavior (Frink & Klimoski, 2004). Personal initiative can be either in-role, for instance, employees can perform their in-role tasks in a proactive way (Frese & Fay, 2001). It can also be extra-role, which is proactive action undertaken beyond the specified means and ends for behavior (Grant & Ashford, 2008), thus, not formally recognized by the organization’s reward system (Hall & Ferris, 2011). This indicates that the forms of personal initiative, which are not perceived to be rewarded, are likely to be avoided. Hence, too much accountability argues that personal initiative will be avoided and concentration is more on the
formal duties for which the employee is held accountable (Mitchell et al., 1998 in Hall & Ferris, 2011).
Certainly, personal initiative is about overcoming difficulties in the pursuit of a goal (Frese & Fay, 2001) and it is suggested as a successful tool to overcome stressors (Fay & Sonnentag, 2002). However, accountability is potentially detrimental in situations where employees believe demands extend beyond their capabilities, revealing the ‘dark side’ (Schlenker et al., 1991 in Hall et al., 2006). Alternatively, high levels of accountability might have positive individual outcomes when in combination with the proper resources, as high autonomy (Hall et al., 2006) or self-efficacy (Royle et al., 2005), to overcome the anxiety that follows from the increased evaluation. However, when the employee does not have or experience these specific resources, high accountability might trigger stress due to the failure of satisfactorily control one’s evaluation process (Hall et al., 2006).
In conclusion, while accountability is essential for the effective running of organizations (Lerner & Tetlock, 1999), employees would also need and perhaps even benefit from some accountability (Hall & Ferris, 2011). Nevertheless, increasing the level of employee accountability would not necessarily result in increased personal initiative. While an employee under moderate accountability will engage in personal initiative because of impression management reasons (Bolino, 2010), under high accountability and lack of proper resources, the evaluative process is seen as a threat, and inducing stress (Hall et al., 2006). This results in a revised second hypothesis regarding the relationship between employee accountability and personal initiative.
H5b: The relationship between task accountability and personal initiative is curvilinear; only moderate levels of accountability will have the expected positive effect.
3 Methods
This methodology section describes how the research is conducted. The research design, the data sample and the instruments used, are thoroughly explained. It continues with the data analysis and preliminary analysis. The section ends with providing the predictions.
3.1 Research Design
The aim of this research is to investigate and expand the current literature on the antecedents of personal initiative of the employee. More specifically, the aim is to examine the positive direct and indirect effect of empowering leadership and team reflexivity and the potential linear or non-linear effect of accountability on personal initiative. A quantitative, multi-source research was carried out by the use of questionnaires in order to measure perceptions of the respondents including behaviors, attitudes and opinions (Saunders, Saunders, Lewis, & Thornhill, 2011). The research was conducted at teams with a leader working in different organizations in the Netherlands. Therefore, two questionnaires were developed, one intended for the team leader, the other for the subordinates, referred to as the employees. Both questionnaires were distributed on two occasions, respectively in the beginning of February 2016 and a second, shorter version, in the beginning of March 2016. By doing so, this research hopes to increase the chance on providing evidence for a causal interpretation of relationships. All questionnaires were available in Dutch and English, established through the comparison of multiple expert translations. The respondents received the questionnaire by email and a web-based questionnaire tool, LimeSurvey, was used to collect the data. Respondents of the same team were matched by their email addresses; therefore this research was not completely anonymous. However, the voluntary nature and strict confidentiality of participation were ensured in the accompanying letter (see Appendix A). In return for their participation, the respondents were offered a report with the main findings and practical implications after finalizing the studies.
The respondents were recruited in collaboration with two other Master students of the University of Amsterdam and the primary thesis supervisor. The requirements for the participating teams were that they consisted of at least three members and their job had to be their main source of income. The respondents were approached by convenience, due to access constraints and time limitations, but also to increase generalizability. This method enabled to achieve an appropriate sample size in a relatively fast and inexpensive way. Therefore selection was based on approaching different individuals or by contacting companies. Other respondents were approached through snowball sampling, where 42 part-time Business Administration Master students were asked to participate with their current team and an additional team of their organization. Because of the non-probability sampling technique and the collection of data by multiple researchers, the sample data is expected to be divers. The respondents are working in a wide variation of sectors, ranging from financial sector, to retail sector, to non-profit sector.
The final data sample consisted of 84 teams, based on 84 leaders and 302 employees. The demographic and organizational variables were measured over 83 employees who also participated during the second round of research (Time 2), establishing that the birth year ranged from 1955 to 1995 (M = 1977; SD = 10.86). Furthermore, more than half of the employees were female (53%) and 37 per cent were male. Tenure of the employees had a wide range from two months up to 33 years (M = 5.44; SD = 5.89). Education levels are divided with the smaller part of the employees (24%) with a degree of MBO or lower, and the larger part (66%) a degree of HBO or higher. A total number of 24 leaders participated in the second research. Leaders ranged in birth year from 1955 to 1992 (M = 1975; SD = 9.18), with 25 per cent female and 75 per cent male. Finally, leaders worked in the organization from 8 months up to 18 years.
3.2 Measures
The four variables were measured from the perspective of the employee. The independent variables (empowering leadership and team reflexivity), the dependent variable (personal initiative) and the mediating variable (accountability) were operationalized using scales adopted from extant literature (see Appendix B). Unless otherwise indicated, all items were measured on a seven-point Likert scale ranging from “completely agree” (1) to “completely
disagree” (7) as endpoint. The questions examined to what extent the employee agrees with a
provided statement about him or herself, about the team or about the leader.
Empowering Leadership measurement was based on the 12-items scale of Ahearne et
al. (2005). It reflects the empowering leadership behaviors as perceived by the employee (α=
.93). The measure has a multi-item subscale comprising of four segments, respectively enhancing the meaningfulness of work, fostering participation in decision making, expressing confidence in high performance and providing autonomy from bureaucratic constraints.
These subscales are distinct, but reflect the overall scale collectively (Zhang & Bartol, 2010).
An example item is ‘My manager helps me understand how my job fits into the bigger
picture’.
Accountability was measured by the eight-item scale developed by Hochwarter et al.
(2003) establishing employee felt accountability (α= .76). The internal consistency and
unidimensionality were demonstrated in prior research (Hall et al., 2003; Hochwarter et al.,
2005). Example of an item is ‘I often have to explain why I do certain things at work’.
Personal initiative was based on the seven-item scale developed by Frese et al., 1997. The scale provides information about the employees’ perception of their personal initiative, related to long-term organizational goals (α= .87). An example of an item is ‘Usually I do more than I am asked to do’.
Team reflexivity was based on the study of Swift and West (1998, in Schippers et al., 2007). A nine-item scale measured the extent to which teams discuss task-related issues, such as evaluating finished tasks, as well as processes, for instance communication (Schippers et al., 2003). An item example is ‘This team often reviews its approach to getting the job done to do a better job next time’. Cronbach’s alpha was .93.
Control variables. The second questionnaire for both employee and leader contained some demographic-related questions, referring to age, gender and tenure, which are all based on a single item. However, since this data was retrieved from only a small sample (N= 70), the variables were inappropriate to serve as control variables. Therefore these variables were not included in further analyses.
3.3 Data Analysis
The descriptive statistics including means, standard deviations, Cronbach’s alphas and correlations of the four variables are provided in Table 4.1. Since the variables could be rated on a scale from 1 to 7, it is indicated that the employees in the sample experience a relatively high level of empowering leadership (M = 5.38, SD = .97) as well as team reflexivity (M = 5.03, SD = 1.10). Additionally, respondents experience a high level of personal initiative (M = 5.74, SD = .68) and rather moderate levels of accountability (M = 4.47, SD = .86).
All scales have a Cronbach’s Alpha above .70, therefore it can be concluded that the scales are sufficiently reliable (Pallant, 2010). Accountability (α = .76) is above the conventional acceptable value although it is not highly reliable. Also, none of the items in the four scales would substantially affect reliability; therefore no items were deleted (Field, 2009). Subsequently, each scale mean was computed by averaging the items, resulting into a variable for empowering leadership, team reflexivity, accountability and personal initiative, to execute the analyses. Preliminary analyses were performed to ensure no violation of the
assumptions of normality, linearity and homoscedasticity. It revealed that the accountability variable is normally distributed, though; the other variables indicate violations of the required assumptions for establishing significance of the Pearson’s correlation coefficient (Field, 2009). To address this problem, the data could be transformed. However, the benefit of transforming the data is utterly debatable (Field, 2009). Moreover, according to the central limit theorem, one can expect that the sampling distribution will be normally distributed, because this research has a large sample (N = 250) and only light tailed distributions within the dependent variable, personal initiative (Wilcox, 2005, in Field, 2009). For that reason the possibly violated assumption of normality are kept in mind, but the variables are not manipulated and it is assumed that the sample is representative for the population.
As indicated before, the employee questionnaires were distributed on two occasions with the goal to increase the evidence for a causal interpretation of the different relationships. However, since the retrieved employee data from both the first and the second measurement is based on a small data sample (N = 70), it is difficult to establish reliable measurements over time. For this reason no analyses will be conducted concerning the second measurement.
3.4 Predictions
Building on the hypotheses as presented before, it is expected to find support for the positive direct relationship between empowering leadership and personal initiative and the positive relationship between team reflexivity and personal initiative. Additionally, this research expects to find support for the mediating, indirect effect of accountability on both direct relationships. Furthermore, it is expected that the effect of accountability on personal initiative is curvilinear, suggesting that the relationship is stronger for moderate levels of accountability than for low or high levels of accountability. When there will be a significant finding for this prediction, the mediating effect of accountability on the two direct
relationships is expected to be curvilinear (inverted U-shaped). The curvilinear relationship is analyzed by linear regression and the mediating relationships are analyzed by PROCESS analysis of Hayes (2012).
4 Results
This sections start with the descriptive statistics. Next, linear regression analysis will establish whether there is a non-linear relationship between accountability and personal initiative. Based on that information the mediating effects will be measured with PROCESS analysis. First, by establishing the mediating effect of accountability on the relationship between empowering leadership and personal initiative. Second, by measuring the mediating effect of accountability on the relationship between team reflexivity and personal initiative.
4.1 Descriptive Statistics
Since PROCESS analysis (Hayes, 2012) follows the approach of excluding cases listwise, the same approach has been used for measuring the correlations. The total sample, on which the mediation analyses were draws, consisted of 250 employees who answered all the investigated variables. IBM’s Statistical Package for Social Sciences (SPSS) automatically detects unanswered questions; therefore no respondents were deleted from the sample. Also, no counter-indicative items were found and no outliers were deleted.
The Pearson’s correlation coefficient in Table 4.1 provides support for the direct relationship between empowering leadership (H1, r = .35, p < .001) and team reflexivity (H3, r = .26, p < .001) personal initiative, establishing support for the first step in the mediation analyses (Field, 2009). Furthermore, it shows that the positive relationship of empowering leadership (H2) and team reflexivity (H4) with accountability is supported, respectively r = .27, p < .001 and r = .19, p < .01. Additionally it indicates a moderate, positive relationship between task accountability and personal initiative, indicating support for H5a (r= .33, p <
.001) with high levels of accountability associated with higher levels of personal initiative. Notably, although not included in the analyses of this research, is a positive high correlation between empowering leadership and team reflexivity (r = .52, p < .001), which will be further elaborated on in the discussion. The expected curvilinear effect (H5b) and the mediating effect of accountability on both direct relationships will be analyzed subsequently.
Table 4.1. Descriptive Statistics and Correlations
M SD 1 2 3 4
1. Empowering Leadership 5.38 .96 (.93) 2. Team Reflexivity 5.03 1.13 .51** (.93)
3. Personal Initiative 5.75 .67 .35** .26** (.88)
4. Accountability 4.47 .86 .27** .19** .33** (.76) ** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed). Listwise N = 250. Cronbach's alpha on diagonal
4.2 Non-linear Relationship
Hierarchical multiple regression was performed to investigate the predictive ability of non-linear accountability to predict levels of personal initiative, after controlling for non-linear accountability. To test the expected curvilinear relationship a new term is created, squared accountability, which is going to represent the non-linear accountability term. To establish whether the model is an accurate representation of the actual data, analyses were performed indicating outliers and residuals. Values for Cook’s distance did not exceed the cut-off maximum value of 1, indicating that outliers will not have a large influence on the regression analysis, therefore no outliers are deleted (Stevens, 2002, in Field, 2009).
Regarding the assumptions for the generalization of the regression analyses, the histogram and scatterplot showed a normal distribution and homoscedasticity for the residuals, and Durbin-Watson value of 1.71 indicated that the residuals are uncorrelated
(Field. 2009). Since the non-linear accountability term is derived from the linear accountability term, resulting in high correlation (r = .99, p < .001), it indicated high multicollinearity and therefore low tolerance for the regression analysis. Nevertheless, this research is interested in indicating the coefficient of determination for the predictive ability of the regression model, which is not affected by multicollinearity; therefore it contains no adverse consequences for the regression analysis (Allison, 2012).
The hierarchical regression analysis included the linear accountability term in the first step (Model 1) and the non-linear accountability term in the second step (Model 2) of the equation. The first model was statistically significant F(1, 248) = 31.15; p < .001 and explained 11,2% of variance in personal initiative. After entry of non-linear accountability in step two, the total variance explained by the model as a whole increased to 11,9%, F(2, 247) = 16.73; p < .001 (Model 2). The Model Summary shows that the introduction of squared accountability in step two, after controlling for linear accountability, accounted for an additional 0.8% variance in personal initiative, although not significantly (R² change = .00; F change (1, 247) = 2.16; p > .05). The results show that non-linear accountability does not have a superior predictive power in this research model. In conclusion, the expected curvilinear relation between accountability and personal initiative where only moderate levels of accountability would have the expected positive effect (H5b) is not supported.
4.3 Mediation
The PROCESS method of Hayes (2012), including bootstrapping (Shrout & Bolger, 2002) allows for testing the mediation of accountability between empowering leadership and personal initiative, and between team reflexivity and personal initiative. Bootstrapping is a non-parametric method for estimating properties of the sampling distribution based on the sample data, and it is an alternative to overcome lack of normality for testing mediation (Field, 2009; Schippers et al., 2008). There are reasons to assume that the indirect effect will
