Transformational leadership and reflexivity and their effects on employee voice quality

(1)

1 Bachelor Thesis: Business Administration

Transformational Leadership and Reflexivity and their

effects on Employee Voice Quality

Student: Koen D. Strijbos Student Number: 10220496

Course: Bachelor Thesis Business Administration Program: Major Business Administration

First supervisor: dr. N. (Nesrien) Abu Ghazaleh Second supervisor: mw. I. (Inge) Wolsink, MSc Organization: University of Amsterdam

Date of submission: 28-06-2015

________________________________________________________________

Abstract

Although proactive behaviour can be of great importance to the success of a company, it depends on the quality of the ideas and suggestions whether it is of any value. To explain when voice quality is best stimulated, I draw on transformational leadership theory. I propose that individualized

consideration and intellectual stimulation (the empowerment dimension of transformational

leadership) lead to higher employee voice quality than can be realized through idealized influence and inspirational leadership (the charismatic dimension of transformational leadership). It is suggested that the first two empower an employee, giving the tools to openly discuss and tackle problems from multiple perspectives, whereas the charismatic leader would make employees more dependent on their leader. More specifically, it is hypothesized that the relation between these dimensions and voice quality is moderated by the self-reflexive behaviours of the leader, as self-reflexivity often leads to higher self-awareness which is the start of a self-improvement process. The hypotheses are tested with samples of 76 triads of managers and employees, each triad consisting of 1 manager and 2 direct subordinates. The hypotheses on the different dimensions were supported, meaning that leaders who show individualized consideration and intellectual stimulation know better how to stimulate employee voice quality than the charismatic leader. Self-reflexivity turned out not to have any effect on

employee voice quality. The findings have important practical implications as they suggest different dimensions of the leadership style may be important in encouraging good voice quality, showing that charisma, which is often associated with great leaders, is not always to be desired.

(2)

2

Statement of originality

This document is written by Student; Koen Strijbos, 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.

(3)

3

Introduction

"After years of telling corporate citizens to 'trust the system', many companies must relearn instead to trust their people and encourage them to use neglected creative capacities in order to tap the most potent economic stimulus of all: idea power." - Rosabeth Moss Kanter (1984) This quote by Rosabeth Moss Kanter, a professor of business at the Harvard Business School, perfectly illustrates how ‘idea power’ in the era of the knowledge society is to be valued by companies and is to be seen as ‘the most potent economic stimulus of all’. As Detert and Burris (2007) put it; employee comments and suggestions intended to improve organizational functioning are critical to performance as it is no longer possible to figure it all out at the top. This phenomenon of upward communication by an employee is known as ‘upward employee voice’; a form of informal extra-role communication by an employee “of ideas, suggestions, concerns, information about problems, or opinions about work-related issues to persons who might be able to take appropriate action, with the intent to bring about improvement or change” (Morrison, 2014, p. 174). Morrison (2014) furthermore discusses two fundamental core assumptions made in literature concerning upward employee voice. The first is that, as the quote also suggested, voice is conceived to be important to organizations and ‘silence’ in fact can be harmful. The second core assumption is that, employees not necessarily share their idea with someone higher up in the organization. As the quote suggests; companies should encourage their employees to excel in their creative capacities. But how can this be achieved? In this piece of research, the importance of good leadership will be looked at and voice empowering leadership qualities will be evaluated.

The fact that upward voicing behaviour is perceived to be of importance to the performance of a company is one of the basic reasons why much research has been oriented towards finding out when employees do and when employees do not voice. According to Detert and Burris (2007) previous research has either focused on individual differences in personality and demographic characteristics of employees, employee attitudes in expressing their discontent (exit, voice and loyalty strategies by Hirschman) or on the organizational context that influence the employees’ willingness to speak up. In line with the last stream of research, they looked at the specific role a leader can have on the willingness of speaking up. Their findings suggest that it is worthwhile investigating to whom someone speaks up and why, they found that: a manager using a transformational leadership style, showing behaviour oriented towards an openness and appreciation for change, encouraged good performing employees to voice more than employees who had a manager who didn’t use a

(5)

5 transformational leadership style (Detert & Burris, 2007). Other research has also shown the impact of the leader on employee voice; Fisk & Friesen (2012) for example found that the perception employees have of the emotion regulation capabilities of their leader influences whether an employee decides to voice or not.

However, an important distinction in literature has been made between voice quantity and voice quality. All previously mentioned studies focused on the actual act of voicing, but, as Wolsink (forthcoming) states: “There is a fundamental difference between having to say something and having something to say” (Wolsink, forthcoming, p. 31), i.e. having to say something does not necessarily mean you have something to say. A study by Shin & Zhou (2003) showed a positive relationship between the transformational leadership style and follower creativity. This finding suggests a positive relationship between transformational leadership and voice quality, however, further research has yet to be done.

Then, what determines good leadership? Particularly few is known on the inherent qualities or practices of a leader that bring about effective leadership, even less is known on how this on turn affects good voice quality of employees. It can however be imagined that a self-reflexive leader, who is aware of his own strengths and weaknesses and actively thinks about his/her actions, who furthermore has the intention to improve his decision making and problem solving, is more likely to accept subordinates’ ideas and opinions. This self-reflexive behaviour, that is expressed by being open to critique, comments and feedback, on turn fosters a psychologically safe environment in which employees more easily come up with ideas, challenging the status quo (Detert & Burris, 2007). Self-awareness is one of several components of authentic leadership, a leadership style that has been positively related to employee voicing behaviour by Hsiung (2012). However, again this research only looked at the effects on voice quantity rather than voice quality.

The focus of this research project therefore is; whether the transformational leadership style has a positive effect on upward employee voice quality and whether this relation is strengthened by a leader’s self-reflexivity. The research question that will be answered is: How does transformational leadership influence voice quality, and how is this relation moderated by a leader’s self-reflexivity? This research question is relevant for practical purposes as good voice quality can lead to better firm performance. The purpose of this study then is to elucidate the role of the leader and his self-reflexive behaviours on voice quality. It will thus lead to practical suggestions to how leaders can improve their employee encouraging behaviours. Moreover, if turns out that self-reflexivity enhances a leader’s effectiveness, it may be a good measure to include in job hiring processes.

(6)

6

Theoretical Framework; Literature review

Upward employee voice; its definition and dimensions

With a rise in global competition, innovation and new product development, there is a

significant challenge for current industry leaders to stay ahead of the competition. Companies nowadays face a much stronger and more difficult environment than back in the old days. Moreover, there is a constant need for companies to bring break-through innovations into the market before their competitors do (Horn & Brem, 2013). In these fast moving circumstances it becomes increasingly important for companies to rely on and stimulate the proactive behaviour of their employees working at the bottom of the pyramid, that is, the employees that are in direct contact with customers. In the course of doing their work and servicing customers, these employees receive, via interaction with those customers, critical information on opportunities for improvement. However, them receiving these potentially valuable pieces of information for matters such as product innovation, product improvement and problem solving, which could be critical to the successful continuation of the company, does not necessarily mean they share this information with someone who can do something with this information. That is, the employee does not necessarily communicate his knowledge or ideas to someone higher in the organization who can take appropriate action with the intent to bring about change.

Where most research focusses on understanding the importance of the act of voicing in contrast to silence (Morrison, 2014; Tourish & Robson, 2006; Detert & Burris, 2007), few research (Wolsink, forthcoming) takes into consideration the content of the message. As such, a distinction can be made between voice action; the amount of input one gives, and voice outcome; the originality and usefulness of the input. In other words, having to say something does not automatically mean you have something to say (Wolsink, forthcoming). In Fortune Magazine, Steve Jobs pointed out the importance of good quality ideas: “Innovation has nothing to do with how many R&D dollars you have. When Apple came up with the Mac, IBM was spending at least 100 times more on R&D. It’s not about money. It’s about the people you have, how you’re led, and how much you get it” (Jobs, 1998). This quote shows the importance of good quality ideas, good employees and good leadership, it may therefore thus be very important to focus on voice quality instead of quantity when doing research in upward employee voice behaviours.

The behaviour of ‘voicing’ is dependent on many different constructs; individual differences in personality, demographic characteristics, coping strategies, and the

(7)

7 organizational context all influence whether someone speaks up or not, and whether the environment fosters the creation of good quality voicing. One of those ‘environmental’ or organizational context factors is leadership style. In the following paragraph the

transformational leadership style will be discussed, especially its effect on voice quality will be studied.

Transformational Leadership; its definition, dimensions and its effect on voice quality

Transformational leadership; the construct

“Leadership can be defined as the reciprocal process of mobilizing, by persons with certain motives and values, various economic, political and other resources, in a context of

competition and conflict, in order to realize goals independently or mutually held by both leaders and followers” (Tourish & Pinnington, 2002, p. 148). Where ‘back in the old days’ leaders would rely on their legitimate or coercive power to force people into following orders, nowadays managers rely on other techniques. Managers often used the ‘transactional

approach’ to persuade people to do as told. It consisted of explaining the subordinate what is required of them and what they would get back in return, often in a monetary sense. Rewards were promised for good performance and corrective action would be taken if employees deviated from the rules (Bass, 1991). However, many times this leads to mediocre performance, as the notion prevails interventions are only needed when things go wrong. Transformational leadership is a leadership style that goes beyond the mere monetary reciprocity and is built around inspiring, energizing, and intellectually stimulating the

employees. It is a leadership style uplifting the moral of the individual, opposed to the simple catering of the employees’ self-interest (Bass, 1999).

The transformational leadership style is characterised by several behaviours that can be classified into idealized influence, inspirational leadership, intellectual stimulation, or individualized consideration (Bass, 1991; Bass, 1999). Idealized influence is a behaviour that creates a bond of trust and confidence, which leads to employees wanting to identify with the leader. In this case, the leader is an example to the employees, who will strive to become like him (Bass, 1991). Inspirational leadership is a dimension close to idealized influence. It is characterized by a leader setting examples and high standard performances as a target that their followers want to reach. However, it is more about creating a challenge using simple words rather than doing so by being charismatic (Bass, 1999). Intellectual stimulation, the third dimension of transformational leadership, comes down to leaders stimulating their

(8)

8 followers to look at problems from multiple perspectives. Teaching them to be more

innovative and creative problem solvers (Bass, 1991; Bass, 1999). A leader displaying individualized consideration is a leader showing coach or mentor-like behaviours. He values the fact that people are inherently different and therefore different in their needs. It is a leader who is personally involved in the development of his employees. (Bass, 1991; Bass, 1999).

While for theoretical and practical reasons these dimensions may be clear (for

practical purposes it may be easier training someone on one of the dimensions instead of all at the same time, especially if the person is already devoting a lot of time into coaching his employees), in quantitative studies the classification has been found in several different constellations (Bass, 1999). Several theorists for example say the first two components (idealized influence and inspirational motivation) both account to charisma whereas the last two (intellectual stimulation and individualized consideration) dimensions are forms of behaviour directed to employee empowerment (Kark, Chen & Shamir, 2003; De Hoogh, Den Hartogh & Koopman, 2004).

In the next section the effects of transformational leadership will be theorized, for doing so the two main dimensions charismatic leadership and empowerment, taken from the ‘Charismatic Leadership in Organizations’ questionnaire (CLIO) byDe Hoogh, Den Hartogh and Koopman (2004), will be considered.

Transformational leadership and its effect on voice quality

According to Morrison (2014) many well-meaning leaders are in fact unintentionally creating an ‘authority-ranking’ social frame that is so prevalent that many employees are afraid to speak up to the ones in power. Upward voice in this sense is a risky situation to the employee, or is at least perceived to be a risky venture. By not realizing this skewed power balance, leaders may have a false sense of reality, as they do not receive the much needed information from the people at the work floor. This in turn might either lead to making wrong decisions or untimely interventions, both with the potential of being destructive to the firm performance. Much research therefore has been focused on when employees do or do not speak up (Detert & Burris, 2007; Fisk & Friesen, 2012; Morrison, 2014). However, as discussed earlier, it is not only voice quantity that matters; voice quality should also be considered. Where Detert and Burris (2007) found that managers using a transformational leadership style had employees who came up with more ideas, it has been argued that creativity and innovation also depend on the work environment people perceive around them (Amabile, Schatzel, Moneta & Kramer, 2003). As in contemporary work environments people often work in

(9)

9 creative teams of professionals, it is the leaders of those teams who direct and evaluate the work and distribute resources and information. In such knowledge-work intensive

organizations, the leader thus is of great importance in facilitating the ‘right’ environment (Amabile et al., 2003). It is then that, since the turn of this century, researchers started focussing on the relationship between transformational leadership style and creativity, of which Shin and Zhou (2003) found a direct positive relationship.

Shin and Zhou (2003) argue that intrinsic motivation is one of the main ingredients for creativity. Being intrinsically motivated means being interested in the excellent completion of a specific task. It is a motivational state in which the employee is energized by working on the task itself, rather than just being interested in the (monetary) outcome, which is characteristic of extrinsic motivation. Whereas the transactional leadership style has been linked to the last, transformational leadership has been linked to the former, that is, the energizing of employees to work on tasks with intrinsic motivation and personal and organizational ‘goal alignment’ (Bass, 1999). This relationships suggests that transformational leadership has an effect on creativity via intrinsic motivation, leading to higher voice quality.

The theoretical argument will now be taken to the separate levels of the two dimensions of the CLIO-questionnaire, charisma and empowerment, as both account for slightly different processes and leadership behaviours, which may not always be looked at through the one-named concept of ‘transformational leadership’. Especially not since Tourish and Pinnington (2002) discuss some thorough critique on the charisma aspect and its effects.

The charisma dimension and its effect on voice quality

The charisma dimension consists of idealized influence and inspirational leadership, both accounting for leadership behaviours that create trust and confidence between leader and follower, an identification with the leader, and the setting of high standards which are believed to be reached by putting in extra effort. All often conveyed through an inspiring manner (Bass, 1991). Shamir, House & Arthur (1993) argue that, by setting these high standards and giving the followers confidence in meeting those standards, they enhance the person’s self-esteem and perceived self-efficacy. Enhanced self-efficacy in turn is believed to be a strong source of motivation. This could mean that, by believing in oneself, having enlarged intrinsic motivation and wanting to live up to the expectations, higher voice quality is reached. Moreover, they pose that charismatic leaders who communicate a strong and vivid vision and mission increase the meaningfulness of and personal commitment to goals and related actions (Shamir, House & Arthur, 1993). This in turn could again lead to a higher

(10)

10 quality of voice. Sosik, Kahai and Avolio (2010) also discuss the importance of intrinsic motivation to creativity. They relate this intrinsic motivational state to transformational leaders encouraging and inspiring their followers, meaning that charismatic leadership could lead to higher voice quality.

Whereas these theories all seem to have a positive feeling about the effects of

charismatic leadership, Kark, Shamir and Chen (2003) discuss the other side of ‘the two faces of charismatic leadership’. Congruent with the just discussed theories they claim that

charismatic leaders can ‘empower’ their followers through self-efficacy raising behaviours; enhancing capacity to think for oneself and to come up with new and creative ideas (Kark, Shamir & Chen, 2003). On the other hand, they suggest charismatic leadership behaviours can create follower dependency on the leader; through the perception that the leader is

extraordinary and exceptional. As Detert and Burris (2007) found in their study, these tendencies might influence the act of voicing as employees will start doubting the quality of their own ideas and suggestions, leaving them with a minimal sense of self-esteem.

The heightened follower dependency is of great concern to Tourish and Pinnington (2002) as well. They note that followers with charismatic leaders need more guidance and have a higher need for leadership than those with non-charismatic leaders. By being inspirational to followers, charismatic leaders have, according to Tourish and Pinnington (2002), the potential to move organizations along the cult continuum and if a leader has enough power to transform the goals of his followers to a more common goal, he could as well be using this power to structure the goals for his own good, they call this ‘the Hitler problem’ (Tourish and Pinnington, 2002). Moreover, it is suggested that charismatic leaders often show narcissist behaviours. As “narcissists tend to be overly sensitive to criticism, can be poor listeners, lack empathy and have a distaste for mentoring and have an intense desire to compete,” they are “inclined to perceive reality through the distorting prism of his or her vision” (Tourish and Pinnington, 2002, p. 152). Having such a confident stance may be overwhelming to followers, leading to less critical feedback and less good quality voicing behaviours.

The empowerment dimension and its effect on voice quality

The second dimension, empowerment, consists of intellectual stimulation and individualized consideration. Both accounting for leadership behaviours that encourage and motivate followers to tackle problems from multiple perspectives in order to come up with creative solutions. This is often done through a direct and personal, coach-like manner (Bass, 1991;

(11)

11 Bass, 1999; Sosik, Kahai & Avolio, 1998). These leaders promote dissent and other ways of thinking, contrary to the charismatic leader as discussed earlier. Through encouraging disagreement and thinking in unusual ways, via a personalized touch of coaching, a

welcoming environment is created in which each and every person is to be valued for his or her ideas. This unique environment fosters situations in which creativity is enhanced and appreciated. Not worrying about being punished for having a different opinion, and by being intrinsically motivated, creativity is enhanced and better voice quality may be the result (Jung, 2001).

Reflexivity; its construct and effects on voice quality

As mentioned earlier, leadership has been looked at from several different perspectives; especially the different leadership styles and their effects on voice quantity and voice quality have been the focus of past research. However, still not much is known about any inherent qualities or clear and observable practices leaders undertake to be effective leaders

stimulating their employees and fostering an environment in which it is okay to challenge the status quo and come up with challenging ideas. Even if leaders want to be open to input from employees, they may still signal otherwise, and may fail to provide adequate mechanisms for participation (Morrison, 2014). It can however be imagined that if a leader is highly aware of his own doings, and thus is very self-reflexive, he may be better able to develop supportive and trusting relationships with his employees. Reflexivity and the effect on voice quality will thus be examined in this research, first the construct will be elaborated on.

Reflexivity can be considered an iterative process in which reflection, planning and action follow up on each other (Widmer, Schipper and West, 2009). This process refers to the considerations of work-related matters and includes behaviours such as: “questioning,

planning, exploratory learning, analysis, diversive exploration, making use of knowledge explicitly, planfulness, learning at a meta-level, reviewing past events with self-awareness, and coming to terms over time with a new awareness” (Widmer et al, 2009, p. 3). This process of reflection is followed up by planning and action, to put the reflection-phase into use and improve work-related practices.

As reflection is an important factor in learning how to become better at doing things, research by Schipper, Den Hartog, Koopman and Knippenberg (2008) and Widmer et al. (2009) focused on the effects of team-reflection and on the antecedents of reflection. They found a positive relationship between transformational leadership style and team reflexivity. A shared vision was encouraged by the inspirational leadership style, which enhanced team

(12)

12 reflexivity and its positive outcomes; heightened forms of innovation, creativity and

effectiveness. Again, this piece of research highlights the importance of a leader to follower performance, specifically the importance of the inspirational leadership style, but moreover, it puts emphasis on the positive effects of team-reflexivity. Reflexivity in this sense is seen as a group-level construct and is defined as “the extent 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” (Widmer et al., 2009, p. 2). As their research focuses on reflexivity as a practice undertaken collectively, this research will focus on whether a transformational leader is better able to develop trusting and supportive relationships with his employees if he himself is reflexively thinking about his own actions.

As mentioned, one of the reflexivity behaviours is self-awareness. Self-awareness is a component of the authentic leadership style, which has been positively related to employee voicing behaviour by Hsiung (2012). However, again this research only looked at the effects on voice quantity rather than voice quality.

Conceptual Models

As discussed in the literature review, the transformational leadership style can either be seen as one ‘all encompassing’ construct, or, can be divided into two separate dimensions; the charismatic and empowerment dimension. As these may all have a different impact on voice quality, these will be looked at separately in the upcoming analysis. Moreover, reflexivity will be taken into account in each separate analysis to formulate hypotheses on the relation

between that leadership style and reflexivity on voice quality.

The relationship between transformational leadership, reflexivity and voice quality

With previously discussed literature in mind the conceptual model for this specific research project can be outlined. The first model that will be looked at is the effect of the

transformational leadership style on voice quality, with reflexivity as the moderating variable. As discussed, it is through the stimulation of intrinsic motivation (Shin & Zhou, 2003) and through goal alignment (Bass, 1999), that a transformational leader fosters a safe

environment for employees to come up with new and creative ideas, and feel valued and secure enough to voice these to the leader. This leads to the following hypotheses:

(13)

13

Hypotheses 1a: The transformational leadership style has a positive effect on voice quality.

In literature, reflexivity shows to be an important factor of learning and improving practices. It can thus be hypothesized that a leader who is constantly re-evaluating his behaviour, and in this way is providing adequate mechanisms for employee participation, is providing their employees with trusting and supportive relationships (Widmer et al., 2009), which in turn may have a positive effect on employee voice quality. Whereas reflexivity has been positively related to group performance on numerous occasions (Schipper, et al,, 2008; Widmer, et al., 2009), a leader’s self-reflexivity hasn’t been linked to voice quality yet.

Hypotheses 1b: Leader self-reflexivity has a direct positive effect on employee voice quality.

It can then also be easily hypothesized that a leader who is showing transformational leadership behaviours which have a positive effect on voice quality, is even more effective in fostering trust and supportive relations when he actively reflects on and tries to improve these behaviours.

Hypotheses 1c: A leader who is high on self-reflexivity enhances the positive

relationship between the transformational leadership style and employee voice quality.

A schematic outline of the hypotheses looks the following:

Figure 1: Conceptual model between Transformation leadership style and employee voice quality.

Transformational Leadership Style Employee Voice Quality Leader Self-Reflexivity

(14)

14

The relationship between the charisma dimension and its effect on voice quality

It is believed that the charismatic dimension of the transformational leadership style is a strong source of intrinsic motivation as it enhances self-efficacy. Furthermore, it is believed that, by creating a strong and vivid vision and mission, the leader increases the

meaningfulness and personal commitment to the goals of his employees (Shamir, House & Arthur, 1993). But, apart from these positive influences of the charisma dimension, Kark, Shamir and Chen (2003) discuss ‘the two faces of charismatic leadership’. They claim that, apart from ‘empowering’ through self-efficacy raising behaviours, charismatic leaders can also create follower ‘dependency’. This in turn might have the effect that employees start doubting the quality of their ideas, leaving them with less self-efficacy than just predicted (Detert & Burris, 2007). Tourish and Pinnington (2002) moreover state that followers with charismatic leaders need more guidance and have a higher need for leadership than those with non-charismatic leaders.

This implies that no clear relationship will be found between the charismatic

dimension of the transformational leadership style as opposing forces are at work, which leads to the following hypotheses:

Hypotheses 2a: No clear relationship will be found between the charismatic

dimension of transformational leadership and voice quality, as opposing forces are at work.

It can however be imagined that a charismatic leader who is fully aware of his actions, and the implications of his behaviours, thus being highly self-reflexive, is creating an

environment open to good voice quality. Tourish and Pinnington (2002) also suggested charismatic leaders often show narcissist traits; being overly sensitive to criticism and having a distorted sense of what is right. Being highly self-reflexive could dampen these narcissist tendencies.

Hypothesis 2b: A charismatic leader who reflects upon his own behaviours and their consequences is fostering an environment in which good voice quality is the result, whereas there is no clear relationship between the charismatic leadership style and voice quality if no reflection takes place.

(15)

15

The relationship between the empowerment dimension and its effect on voice quality

The other dimension of transformational leadership, empowerment, is believed to promote diversity and unconventional thinking, which creates an environment in which different ideas are appreciated and stimulated, leading to higher forms of creativity (Jung, 2001).

Furthermore, people are intrinsically motivated through their direct and personal contact with the leader, who shows coach-like behaviours (Bass, 1991; Sosik, Kahai & Avolio, 1998). These propositions lead to the following hypothesis:

Hypothesis 3a: The empowerment dimension of the transformational leadership style has a direct positive effect on employee voice quality.

A leader who is also highly self-reflexive would lead to the following hypothesis:

Hypothesis 3b: A leader highly aware of the positive behaviours and outcomes of his ‘empowerment’ style, through a heightened self-reflexivity, would logically be able to use more appropriate mechanisms to foster voice quality than a leader who is low on self-reflexivity.

Research Design and Methodology

Data

The level of analysis of this empirical research is the individuals inside the organization. Especially the leader and his followers were the focus of this study. To be able to find correlations and relations between a leadership style and the follower’s voice quality, a quantitative research design has been used as quantitative data collection is excellent for testing the theory with a large amount of people (Field, 2013). An existing dataset with previously described variables is present and has been used for the purpose of this study. It is based on two different surveys, one filled in by the leader and one by his two followers (his subordinates). In total 76 triads, consisting of 1 leader and 2 subordinates, completed the survey in the presence of one of the researchers. The duration of the survey was

approximately between 10 and 15 minutes for the supervisor, the questionnaire the employees had to fill in took between 20 and 25 minutes to finish. All participants were Dutch speaking residents selected from the personal networks of the researchers, it is thus a convenience

(16)

16 sample that has its limits in generalizability. Due to the time-limit it was not possible to actively look for people from all layers in society. However, as these triads were selected by 76 different researchers (as part of a research project at the University of Amsterdam), a wide range of jobs, industries and sectors is included in the sample.

Analyses and Predictions

This study proposed that managers who show transformational leadership behaviours will have a positive effect on the voice quality of their employees (H1a). As it was argued that transformational leadership could be theoretically separated into two different dimensions, each with their own implications for employee voice quality, it was predicted that: no clear relationship will be found between the charismatic dimension of transformational leadership and employee voice quality as several opposing forces are at work (H2a), whereas a positive relation is expected for the empowerment dimension of transformational leadership on employee voice quality (H3a). First a factor analysis will be carried out to check for the existence of these two dimensions. Furthermore, it was hypothesized that a manager who regularly reflects upon his work-related behaviours creates an environment that fosters employee voice quality (H1b). It is these variables that will be looked at in the first model of analyses, with a simple regression model in SPSS. In the second model of analyses, the moderating effect of the reflection variable will be will be looked at to test the predicted interaction effect, which stated that in each situation described in H1a, H2a and H3a the effect of transformational leadership on employee voice quality will be higher when the leader regularly reflects upon his work-related behaviours (H1c; H2b; H3b). Again a regression analyses will be done in SPSS and the first model will be compared to the second to see which of the two has more predictive power.

Operationalisations of the main variables:

The dependent variable Voice Quality

The dependent variable voice quality, rated by the manager, is based on a 26-item scale (α = .96) developed by Wolsink (forthcoming), focussing on the originality and utility of ideas (e.g. “Ideas from [name employee] are innovative,” “are practically not feasible”). All voice quality questions are based on a 7-point likert scale ranging from (1) does not apply at all to (7) completely applies. Any counterbalanced items were recoded for further analysis. After conducting a Principal Components Analysis with varimax rotation, of which a more detailed

(17)

17 description can be found in the appendix, only 17 items remained (α = .97), 9 items were dropped due to low communality and cross-loadings. The Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO = .95 (‘marvellous’; Field, 2014). The PCA had the originality (10 items, α = .97) and usefulness (7 items, α = .89) factors

confirmed, with a combined explained variance of 72.70%. A mean out of the 17 variables is created to reflect the dependent variable of this study; Voice Quality.

The independent variable; leadership style:

Transformational leadership is the independent variable in this study. It is measured on a multi-source basis; both subordinates had to fill in questions concerning the leadership style of their manager. It is based on a 12-time scale (α = .95) adapted from De Hoogh, Den Hartogh & Koopman (2004) (e.g. “My leader stimulates subordinates to think independently and creatively,” “has a clear vision and an image of the future,” and “gives subordinates a sense of working for an important and good cause”). The scale has been used and has been validated in several previous studies on leadership style (Den Hartog & Belschak, 2012). After conducting a Principal Components Analysis (PCA) only 10 of the original 12 items remained (α = .95), as 2 of the 12 items had communalities below 0.5. The Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO = .94 (‘marvellous’; Field, 2014). A more detailed description of the PCA can be found in the appendix.

As previously discussed, transformational leadership consists of several different dimensions (idealized influence, inspirational leadership, intellectual stimulation, or individualized consideration), all touching on different leadership characteristics. These different dimensions therefore all may have very different consequences; a two dimensional forced factor analysis has been carried out to distinguish charismatic leadership, consisting of 4 items (α =.86), from empowerment leadership, based on 6 items (α = .93), as has been done in various previous studies (Kark, Chen & Shamir, 2003; De Hoogh, Den Hartogh &

Koopman, 2004). Centered variants of the variables have been used in the regression analysis.

The moderating variable self-reflexivity:

The moderating variable in this study is a leader’s self-reflexivity. It is a 6-item based (α = .71) questionnaire based on a 5-point likert scale ranging from (1) never to (5) often (e.g. “I critically asses the information and methods I use to make decisions” and “I integrate

(18)

18 constructive feedback from others into a new way of working”). The reflexivity measure used in this study is based on the six item reflexivity measure of Schippers which has been used in several previous studies (Schipper, Den Hartog, Koopman & Knippenberg, 2008; Widmer, Schipper and West, 2009). However, as their measure is based on the group-level reflexivity construct, questions such as ‘We regularly reflect on the way in which we communicate’ are slightly changed to reflect an individualized measure of reflexivity: “I evaluate whether information exchange, concerning work-related topics, with my fellow employees develop at a satisfactory level”. The centered variable has been used in the regression analysis.

Operationalisations of the control variables:

Gender

As female leaders in general tend to show more transformational leadership behaviours than their male counterparts (Bass, 1999), the leader being male or female may matter to the relationship between transformational leadership and voice quality. Gender will therefore be included as a control variable. It is included as ‘I am a …’ with (1) is man and (2) is woman.

Positive Affect

Positive affect is taken into consideration as someone’s mood at the time of performance appraisal can have an effect on the performance appraisal outcome, that is; mood can be the cause of rating distortions (Sinclari, 1988; Robbins & DeNisi, 1998). As voice quality is manager rated, it is important to take the manager’s mood at the time of making the appraisal (at the time of filling in the survey) into account. The literature is divided about the effect of Negative Affect, after having looked at this variable in a correlation table, it has not been valued as a control variable for this piece of research. Positive affect however did show correlations with several of the main variables. It has been measured by 6 separate mood questions: “To what extent do you feel calm/serene/relaxed/happy/elated/upbeat?”, based on a 5-point likert scale (α = .80; 1 = to a very low extent, 5 = to a very high extent).

(19)

19

The Analysis

Sample descriptive statistics

As stated before, the dataset consisted of 76 Dutch speaking teams, consisting of 2

subordinates and 1 manager. The employees answered several questions about their selves, their colleagues and their manager. The managers answered questions about themselves and their two subordinates. Of the managers, 46 are male (65%) and 25 female (35%) female, the average age is 42.17 years old (SD = 11.87), on average they had 134.21 months (SD =

126.58) months of service. Of the employees, 73 are male (50%) and 73 are female (50%), the average age is 32.89 (SD = 12.13) (after deleting an age of 4), and they have an average tenure of 75.96 months (SD = 98.22). Overall, the most common sector the respondents are in is the private sector (45%), followed by the retail (15%) sector, the entire distribution can be found in table 1.

Table 1: Distribution of sectors respondents are in

Sector Non-profit Government University Private Care Retail Owner Other

N (%) 18 (8%) 10 (4%) 14 (6%) 101 (45%) 16 (7%) 35 (15%) 6 (3%) 26 (12%)

Reliabilities and Correlations

As already mentioned, to see whether all measurements can be used for the regression

analysis they have been internally tested with a reliability analysis. All turned out to be higher than 0.7, which is satisfactory (Field, 2014). In Table 2 all Cronbach Alpha’s are reflected on the diagonal. Furthermore, Table 2 contains the correlations between the main variables and some control variables. As expected, transformational leadership and empowerment

leadership correlate with voice quality, r(141) = .18 and r(141) = .22 respectively. Also, as predicted, no correlation between charismatic leadership and voice quality exists r(141) = .07, ns. Furthermore, against predictions, reflexivity only marginally correlates with voice quality r(141) = .15. An interesting outcome however is that reflexivity shows to correlate with gender and positive affect, r(141) = .19 and r(141) = .35. Meaning that gender and positive affect somehow influence the self-reporting reflexivity measurement.

(20)

20

Table 2: Descriptives and correlations between the variables (Cronbach’s Alphas on diagonal; N = 141) Variables M SD 1 2 3 4 5 6 7 8 1. Voice Quality 4.26 1.16 (0.97) 2. Gender 0.35 0.48 -.087 (1.00) 3. Positive Affect 3.04 0.70 ,278** _-.003 _(0.80) 4. Negative Affect 1.32 0.48 .059 -,195* _-.159+ _(0.88) 5. Transformational Leadership 5.04 1.25 ,182* _.128 _.116 _.019 _(0.95) 6. Charismatic Leadership 5.18 1.21 .065 .125 .090 .028 ,882** _(0.86) 7. Empowerment Leadership 4.98 1.35 ,218** _.116 _.107 _.011 _,974** _,766** _(0.93) 8. Reflexivity M 3.85 0.46 .149+ _,188* _,353** _.049 _.059 _.048 _.049 _(0.71)

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

+_{. Correlation is marginally significant at the 0.1 level (2-tailed).}

Results

The relationship between transformational leadership and voice quality

To examine the relationship between transformational leadership and voice quality, a linear regression analysis has been conducted in SPSS. The first model shows the relation between the main variables, transformational leadership, reflexivity and voice quality, and the control variables. As expected, transformational leadership shows a positive regression with voice quality (β =.176, p< .05), confirming H1a. However, against expectations the model shows no regression between reflexivity and voice quality, rejected hypothesis H1b that stated that reflexive leaders would positively influence the voice quality of their employees. As the correlation table already suggested, a regression does exist between positive affect and voice quality, which would mean that the better the mood the rater is in, the better the performance appraisal. The first model has an explained variance of 11.9 percent. All coefficients can be found in Table 3.

In the second model the interaction term was added to test for moderation effects of reflexivity on the relation between transformational leadership and voice quality. Against

(21)

21 expectations no regression has been found, rendering the second model with no more

explanatory power than the first; leaving H1c, that stated that self-reflexive transformational leaders would raise employee voice quality more than non-self-reflexive transformational leaders, unconfirmed.

Table 3: Results regression analysis between TL, Reflexivity and Voice Quality

Voice Quality (DV) Model 1 Model 2

Coefficient SE Beta Coefficient SE Beta

Constant 3.231*** .458 3.227*** .458 Gender (Manager) -.317 .200 -.131 -.285 .203 -.118 Positive Affect .376** .143 .229 .376** .143 .229 Reflexivity Transformational Leadership Transformational*Reflexivity . 206 .157* .222 .077 .082 .176 .196 .150* -.164 .222 .077 .178 .078 .161 -.075 R2 _,119** _,125

Note: Dependent variable = Voice Quality, N= 141 *=p < .05. ** p<.01. ***p<.001. +_{= marginally significant}

The relationship between charismatic leadership and voice quality

To examine the relationship between charismatic leadership and voice quality, another linear regression analysis has been conducted in SPSS. The first model shows the relation between the main variables, charismatic leadership, reflexivity and voice quality, also the control variables are included. As expected, charismatic leadership shows no regression with voice quality, confirming H2a which stated that no clear relationship between charismatic

leadership and voice quality would be found. The first model has an explained variance of 9.5 percent. All coefficients can be found in Table 4.

The second model consisted the interaction term to reflect the effect of reflexivity on the relation between charismatic leadership and voice quality. Against expectations no regression has been found, rendering the second model with no more explanatory power than the first, leaving H2b, that stated that a charismatic leader would have a positive effect on voice quality is he was highly self-reflexive, unconfirmed.

(22)

22

Table 4: Results regression analysis between Charismatic Leadership, Reflexivity and Voice Quality

Constant 3.144*** .463 3.140*** .462 Gender (Manager) -.281 .203 -.116 -.241 .205 -.100 Positive Affect .401** .145 .245 .400** .144 .244 Reflexivity Charismatic Leadership Charismatic*Reflexivity .203 .052 .225 .080 .081 .053 .180 .050 -.233 .225 .080 .191 .072 .051 -.101 R2 _,095** _,010

Note: Dependent variable = Voice Quality, N= 141 *=p < .05. ** p<.01. ***p<.001. +_{= marginally significant}

The relationship between empowerment leadership and voice quality

To examine the relationship between empowerment leadership and voice quality, a third linear regression analysis has been done in SPSS. The first model shows the relation between the main variables, empowerment leadership, reflexivity and voice quality, also the control variables were added. As expected, empowerment leadership shows a regression with voice quality, (β =.18, p< .01), confirming H3a. The first model has an explained variance of 13.3 percent. All coefficients can be found in Table 5.

In the second model the interaction term was added to check for an interaction effect of reflexivity on the relation between empowerment leadership and voice quality. Against expectations no regression has been found, rendering the second model with no more

explanatory power than the first, leaving H3b, which stated that a self-reflexive empowerment leader would have a more positive effect on voice quality than a non-self-reflexive

empowerment leader, unconfirmed.

Table 5: Results regression analysis between Empowerment Leadership, Reflexivity and Voice Quality

Constant 3.246*** .454 3.243*** .455 Gender (Manager) -.324+ _.198 _-.134 _-.303 _.201 _-.125 Positive Affect .372** .142 .227 .371** .142 .226 Reflexivity Empowerment Leadership Empowerment*Reflexivity .209 .179** .220 .070 .083 .205 .206 .175* -.109 .220 .071 .160 .082 .201 -.055 R2 _,133** _,136

(23)

23

Discussion

This study aimed to contribute to the literature on employee voicing behaviour, as voice can be of great importance to strategic decisions. Of interest was the moderating effect of a leader’s self-reflexivity on the relation between leadership style and employee voice quality, which resulted in the following research question: How does transformational leadership influence voice quality, and how is this relation moderated by a leader’s self-reflexivity?

Hypotheses and key findings

Overall, the results were quite unexpected. However, support was found for the first hypothesis, saying that the transformational leadership style has a direct positive effect on employee voice quality. This means that when a leader shows behaviours that go with the transformational leadership style, employees tend to voice higher quality ideas. This finding adds to the study by Shin and Zhou (2003), showing that the transformational leadership style not only heightens one’s creativity, but also leads to higher voice quality, consisting of both originality and utility. This would suggest an association between creativity and voice quality, as has also been partly argued by Wolsink (forthcoming).

The hypothesis that stated that leader self-reflexivity has a positive effect on employee voice quality was rejected. It turned out that the two were not associated with each other, which suggests that a leader being highly self-reflexive in his work-related behaviours does not necessarily stimulate his followers to show higher voice quality. A plausible explanation for this finding could be that higher self-reflexivity and self-awareness may lead leaders to “become more transparent in communicating their values, identity, emotions, goals and motives to others” (Gardner, Avolio, Luthans, May & Walumbwa, 2005, p. 358). As this coincides with behaviours from the charismatic leadership style, which also does not seem to have a positive effect on voice quality, it could very well be that reflexivity does not associate with voice quality. However, it should be noted that positive affect did have an effect on both voice quality and reflexivity. Suggesting that people in a more positive mood, at the time of completing the surveys, more highly evaluate their followers on voice quality and more positively assess their own self-reflexive behaviours. These performance appraisal and rating distortions have been found in several research projects before and should be considered in any future studies (Sinclari, 1988; Robbins & DeNisi, 1998).

Then, also unexpected, hypothesis 1c, that stated that a leader’s self-reflexive behaviours would affect the relation between transformational leadership and voice quality,

(24)

24 also has been rejected. This suggests that managers who show transformational leadership behaviours and are highly self-reflexive do not necessarily stimulate higher voice quality than transformational leaders who are not self-reflexive.

As just stated, a direct association between the charismatic leadership style and voice quality has not been found. This is in line with hypotheses 2a that proposed that no clear relation would be found between the charismatic dimension of transformational leadership and voice quality. In agreement with Tourish and Pinnington (2002), this suggests that opposing forces are at work, both empowering and disempowering, when it comes down to a charismatic leader.

The leader’s self-reflexivity again was not influencing this relation in any way, as was suggested earlier that self-reflexive behaviours may instil some of the same behaviours that go with the transformational leadership style (Gardner et al., 2005), thereby rejecting hypothesis 2b.

As predicted, hypothesis 3a is supported, which stated that the empowerment dimension of the transformational leadership style has a direct positive effect on employee voice quality. This puts forward that employees having transformational leaders, who emanate empowering characteristics, show higher quality voicing behaviours than employees who have leaders who do not show these empowering behaviours; coaching, individual

consideration, etc. This is in line with research done by Sosik et al. (1998) and Jung (2001), who say employees are intrinsically motivated through direct and personal contact with the leader. Again, Jung (2001) only spoke of a positive association between empowerment and creativity. Therefore, this study is adding to the literature the positive effect of the

empowerment dimension on voice quality. Once more, self-reflexivity did not have an effect on this relationship, rendering the hypothesis that stated that a highly self-reflexive leader enacting empowerment behaviours would be able to use more appropriate mechanisms to stimulate voice quality than an ‘empowerment leader’ who is non self-reflexive (H3b), rejected.

Critique and suggestions for future research

Several limitations of this study should be considered and will be discussed. A

methodological critique point of this study may be that a single-source measure has been used for assessing voice quality. As this study discussed upward voice quality to the manager, it was only possible to let the manager assess the quality of the ideas and suggestions of his

(25)

25 employees. Therefore it was open to error and bias by the manager, as was seen and discussed previously through the effect of positive affect on the voice quality variable. Moreover, it could be argued that when the manager feels threatened by the suggestions, or the ideas are not ‘up his sleeve’ he will value them accordingly, thus not being objective (Tourish & Pinnington, 2002). For this piece of research it was impossible to evaluate upward voice quality in any other way. However, in many cases the manager will be the person to decide whether to use the ideas and suggestions or not, therefore it does make sense to use a single-source measure. For future research however, it is suggested to take into account any written materials that employees hand their managers, and let at least two researchers evaluate these ideas and suggestions, in order to get more objective results. An experiment concerning voice quality and leadership behaviours could also be realized. Here you would first let several people assess whether a leader shows charismatic or empowerment behaviours. Then, these leaders could each be assigned a group of people in which the leaders assign a task and voice quality is later assessed by independent outsiders (two or more researchers). However, in this case one would probably measure creativity stimulation by leadership style instead of voice quality, which again shows the difficulty of this specific piece of research.

The same problems are encountered when looking at the reflexivity measure, which also is a single-source based measure, relying on the self-assessment of the manager. As this measure has not been used in this way before, it is a new scale that should be tested further in upcoming research. As positive effect and reflexivity highly correlated with each other, it may very well be that no hypotheses regarding reflexivity was supported because it needs to be measured in a different way. It can for example be measured at multiple times, trying to minimize the effect of mood (positive affect), on the self-assessment.

Another critique point could be that the questioned managers do not know much about this subject, and may therefore wrongly interpret the questions concerning voice quality, accidentally mixing it up with for example voice quantity. To ward off any misinterpretations, the managers were asked control questions, such as: The questions I am about to answer are related to ‘how many times my employee comes up with an idea’ or ‘the quality of the ideas my employee comes up with’, these questions were accompanied by a brief explanation on the subject.

Yet, it could also be argued that the data does not represent the entire Dutch

population and has limited generalizability power, as it has been obtained via convenience-sampling (Field, 2014). Whereas at first sight this may be true, the data was obtained by 76 different students, resulting in a fairly well representation of all sectors. Though it may be

(26)

26 argued that most of the respondents are from the Randstad, these kind of companies may well be the kind of focus you are looking for in such a research project. Moreover, the distribution of men versus women holding a managerial position in this dataset is very much skewed, in favour of the men (65%), which, for such a sample size, is a good reflection of the reality (79%) (Sools, Baerveldt & van Engen, 2011). Also, the distribution of men and women in the Dutch working population is 55% to 45%, this comes close to the distribution in the data, which is even (CBS, 2015).

As many previous studies have focussed on employee aspects, or on visible leadership behaviours (such as leadership style or leadership emotions), future research could dive more deeply into the inherent characteristics and behaviours of the leader that result in good voice quality. I would suggest to, once more, more deeply research the self-reflexivity measure, especially on work-related matters. Ruling out the effect of the mood a participant is in would be a good starting point in making the reflexivity measure more robust. Moreover, it is clear that much more future research should focus on the quality aspect of voice, instead of the quantity aspect. A qualitative case study could give deeper insight into some of the ongoing processes, which could set the base for more and new quantitative testing.

Contributions

Despite several limitations and unexpected results, this study clearly had its contributions. By looking at voice quality instead of voice quantity, the outcome of voice received more

theoretical attention than the mere act of voicing, which has been the focus of much previous research. As still very few is known about the actual process of voicing, this study showed that future research could focus more on questions such as ‘when is good quality voice stimulated’ instead of questions such as ‘when does someone voice’ or ‘when does someone decide not to voice’. Not only will this be important for theory building, also for practical purposes it can be of great value. Whereas charismatic leaders often are seen as great leaders, setting the example and being valuable to a company, it is now shown that charismatic leaders actually constrain the flow of good quality suggestions and ideas from employees to managers higher up in an organization. Especially in times of fast moving environments, in which first-hand information from people working at the bottom of the pyramid can be crucial to a company’s success, charismatic leaders could pose a threat to the company’s survival. Instead, it would be wise to look for candidates that show individual consideration, and intellectual stimulation, thereby empowering employees to come up with good quality ideas and suggestions. Moreover, by breaking down transformational leadership into different

(27)

27 dimensions, this study also added to practical insights into the training and development of leaders. By breaking it down into these dimensions, it becomes much clearer for leaders to understand on what dimension they could excel themselves, to become better leaders.

Conclusion

In a fast moving economy it can be of great importance to quickly hop on the boat of new trends and technologies. Spotting these on time can have great effect on the success of a company. However, it does not end there. Ideas and suggestions should be communicated to someone who can bring about change, and who can do something with the idea. In this case, it is best if people at the top receive good quality suggestions which are useful and original, and can be put to practice instantaneously. An environment that stimulates good quality voice is set by the transformational leader, especially one who knows how to empower his employees. Charismatic leaders on the other hand tend to make their employees dependant on them, not encouraging their employees to come up with good quality ideas and suggestions. A leader’s self-reflexivity does not seem to be of any importance and does not stimulate or dampen employee voice quality in any way.

(28)

28

References:

Amabile, T. M., Schatzel, E. A., Moneta, G. B., & Kramer, S. J. (2004). Leader behaviors and the work environment for creativity: Perceived leader support. The Leadership

Quarterly, 15(1), 5-32.

Centraal Bureau voor de Statistiek (2015). Arbeidsdeelname; Kerncijfers. Den Haag, Heerlen. Retrieved on June 23rd, 2015, from:

http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=71738ned&D1=0-6,22-23&D2=a&D3=a&D4=0&D5=l&HDR=T&STB=G4,G1,G2,G3&VW=T

De Hoogh, A. H. B., Den Hartog, D. N., & Koopman, P. L. (2004). De ontwikkeling van de CLIO: een vragenlijst voor charismatisch leiderschap in organisaties. Gedrag en Organisatie, 17(5), 354-381.

Den Hartog, D. N., & Belschak, F. D. (2012). When does transformational leadership enhance employee proactive behavior? The role of autonomy and role breadth self-efficacy. Journal of Applied Psychology, 97(1), 194.

Detert, J. R., & Burris, E. R. (2007). Leadership behavior and employee voice: Is the door really open?. Academy of Management Journal, 50(4), 869-884.

Field, A. (2013). Discovering statistics using IBM SPSS statistics. Londen: Sage

Fisk, G. M., & Friesen, J. P. (2012). Perceptions of leader emotion regulation and LMX as predictors of followers' job satisfaction and organizational citizenship behaviors. The Leadership Quarterly, 23(1), 1-12.

Gardner, W. L., Avolio, B. J., Luthans, F., May, D. R., & Walumbwa, F. (2005). “Can you see the real me?” A self-based model of authentic leader and follower development. The Leadership Quarterly, 16(3), 343-372.

Horn, C., & Brem, A. (2013). Strategic directions on innovation management-a conceptual framework. Management research review, 36(10), 939-954.

Hsiung, H. H. (2012). Authentic leadership and employee voice behavior: A multi-level psychological process. Journal of business ethics, 107(3), 349-361.

Jung, D. I. (2001). Transformational and transactional leadership and their effects on creativity in groups. Creativity Research Journal, 13(2), 185-195.

Kanter, R. M. (1984). Recommended for Summer Reading: Heralding a Renaissance for Corporate America—Relying on the People. Change: The Magazine of Higher Learning, 16(5), 18-47.

Kark, R., Shamir, B., & Chen, G. (2003). The two faces of transformational leadership: empowerment and dependency. Journal of applied psychology, 88(2), 246.

Morrison, E. W. (2014). Employee Voice and Silence. Annu. Rev. Organ. Psychol. Organ. Behav., 1(1), 173-197.

(29)

29 Robbins, T. L., & DeNisi, A. S. (1998). Mood vs. interpersonal affect: Identifying process and

rating distortions in performance appraisal. Journal of Business and Psychology, 12(3), 313-325.

Schippers, M. C., Den Hartog, D. N., Koopman, P. L., & van Knippenberg, D. (2008). The role of transformational leadership in enhancing team reflexivity. Human Relations, 61(11), 1593-1616.

Shin, S. J., & Zhou, J. (2003). Transformational leadership, conservation, and creativity: Evidence from Korea. Academy of management Journal, 46(6), 703-714.

Sinclair, R. C. (1988). Mood, categorization breadth, and performance appraisal: The effects of order of information acquisition and affective state on halo, accuracy, information retrieval, and evaluations. Organizational Behavior and Human Decision Processes, 42(1), 22-46.

Sools, A., Baerveldt, C., & van Engen, M. (2011). De paradox van ambitie. Een

cultuurpsychologische analyse van het vertoog over hogerop komen als manager in een multinational in Nederland vanuit een genderperspectief. Tijdschrift voor Genderstudies, 5(2).

Sosik, J. J., Kahai, S. S., & Avolio, B. J. (1998). Transformational leadership and dimensions of creativity: Motivating idea generation in computer-mediated groups. Creativity Research

Tourish, D., & Pinnington, A. (2002). Transformational leadership, corporate cultism and the spirituality paradigm: An unholy trinity in the workplace?. Human Relations, 55(2), 147-172.

Tourish, D., & Robson, P. (2006). Sensemaking and the distortion of critical upward communication in organizations. Journal of Management Studies, 43(4), 711-730. Widmer, P. S., Schippers, M. C., & West, M. A. (2009). Recent developments in reflexivity

research: A review. Psychology of Everyday Activity, 2(2), 2-11. Wolsink, I. (forthcoming). Attention Control and Voice Quality.

(30)

30

Appendix: Principal Component Analysis; description of steps taken

Voice quality

A principal factor analysis was conducted on the 26 items with Varimax rotation, based on eigenvalues greater than 1. After initial analysis, 4 factors are extracted. Items 12, 14 and 22 all have a communality below 0.5. Moreover, the counterbalanced items focusing on

originality (2, 9 and 12) all load on different factors, whereas you would expect them to load on the same. It can thus be concluded that these questions have been misinterpreted, therefore they are taken out of the PCA.

Now, 3 factors remain and all communalities are above 0.5. One of these factors only consists of the CB-items, it can thus be concluded that these items are misinterpreted, the PCA is ran again without these two items (17, 26).

Now only 2 factors remain; item 16 has low communality and item 15 shows cross-loading. The PCA is ran again without these two items.

The remaining 2 factors have communalities above 0.5, and show that voice quality can be divided in originality and utility. Originality consists of 10 items (α = .97) and usefulness of 7 items (α = .89), combined they have an explained variance of 72.70%.

Items upward employee voice quality (α = .96)

Originality (α = .97)

Factor loading Communality

1. Als mijn werknemer met een idee komt, is dit origineel.

,856 ,833

,542 (and ,527) ,600

3. Mijn werknemer heeft een verfrissende mening.

,790 ,784

4. De suggesties van mijn werknemer zijn origineel.

,872 ,830

5. Als mijn werknemer een idee aandraagt, is dit vernieuwend.

,835 ,822

6. Mijn werknemer komt met originele adviezen.

,844 ,814

7. De ideeën van mijn werknemer zijn innovatief.

,886 ,864

8. Als mijn werknemer een advies geeft, is dit verfrissend.

,750 ,765

,813 ,732

10. De mening van mijn werknemer is origineel.

Transformational leadership and reflexivity and their effects on employee voice quality