The Effects of Transformational Leadership, Behavioural Activation, and Creativity on Voice Quantity and Voice Quality
MSc. in Business Studies – Leadership and Management Track Adeline Dessing
University of Amsterdam August 26th 2014
Master thesis under supervision of: Inge Wolsink and Wendelien van Eerde
Word Count: ± 8212 (excl. references and abstract) Student number: 10664785
Abstract
Not much research has been done on the antecedents of voice quantity and voice quality. The following research question was investigated in this paper: What is the influence of
transformational leadership on the relationship between behavioural activation and voice quantity? How does creativity affect the relationship between voice quantity and voice quality? And is the relationship between behavioural activation and voice quality mediated by voice quantity? A survey study was conducted to investigate this. Data was collected among triads (N = 76) which consisted of one supervisor and two employees. We investigated whether there was an interaction effect between voice quantity and creativity on voice quality. Secondly it was investigated if there was an interaction effect of behavioural activation and transformational leadership on voice quantity. We found an interaction effect of voice quantity and creativity on voice quality. Secondly a negative relation was found between behavioural activation and voice quality. We conclude that there might be two antecedents for voice quality; creativity and high voice quantity.
Keywords: voice quantity, voice quality, behavioural activation, transformational leadership, creativity
‘Communication - the human connection - is the key to personal and career success’ Paul J. Meyer
Paul J. Meyer has dedicated his live to motivating people to work at their full potential. He believed that by implementing the right leadership and motivation, people could achieve success for themselves and their work environment (https://www.pauljmeyer.com). Besides motivation and leadership, communication is also an important tool for gaining success.
Increasingly, companies these days are gaining their success by embracing open communication in their companies. Open communication cultures are known to have a positive influence on creativity and innovation (Hassel, 2013; Martins & Terblanche, 2003). In an open
communication culture, the active sharing of non-confidential and non-proprietary information among employees and leaders is encouraged (Fagan-Smith, 2011). When a person is actively engaging in this open communication culture by constructively talking about ideas, suggestions, or opinions, which can be used to make positive changes for the organization, this person is voicing (Van Dyne & LePine, 1998). By voicing ideas, employees can contribute to the performance of a company (Morrison, 2011; Liu, Zhu & Yang, 2010). Therefore it seems important to know what the drivers are for initiating voice behaviour and how they translate into qualitative voice that benefits the company.
Voice behaviour has not been extensively researched academically. Most research regarding voice behaviour has been done towards explaining what the motives are for how often employees voice, which can be referred to as voice quantity (Detert & Burris, 2007; Liu et al., 2010). Not much attention has yet been paid to the outcome of voice behaviour. This entails the actual usefulness and originality of the ideas voiced by an employee. Wolsink (forthcoming) has
extended the concept of voice behaviour. She divided voice behaviour into two constructs. Voice quantity is about how often people actually voice their behaviour. The second one, voice quality is about whether the voiced ideas are original or useful. Wolsink (forthcoming) stated in her study that besides voice quality and quantity having different outcomes, the processes preceding these behaviours also differ from each other. For both theory and practice it seems of value to investigate what the preceding processes are that lead to voice quantity and voice quality. These ideas could be of value for companies that are involved in the generation of ideas. We expect to attribute the preceding processes of voice quantity to personal characteristics. In order for one to voice ideas of high quality we expect that certain abilities are needed to create these ideas.
Detert and Burris (2007) state that personal characteristics can be an influencing factor on voice quantity. In this study we focus on two types of personal characteristics. Firstly we will focus on behavioural activation in relation to voice quantity. Persons with behavioural activation engage in more goal-directed behaviour and will experience positive feelings when being
exposed to impending rewards (Carver & White, 1994). We want to examine if behavioural activation will cause one to voice more often. Secondly we will focus on creativity in relation to voice quality. We expect that creative people are able to voice high quality ideas.
Another factor of influence could be the organizational context of a company (Detert & Burris, 2007). An organizational context in this matter could be the kind of leadership shown by supervisor towards employees. Liu et al. (2010) have shown in their study that transformational leadership facilitates the process of voice behaviour. This means that the presence of a
transformational leader causes employees to voice more often. In this study we will focus on the influence of transformational leadership.
In short, we will investigate how leadership, individual levels of behavioural activation, and creativity dependently and interactively influence how often people voice, and what (in terms of quality) people voice.
The practical relevance of this study is twofold. Firstly, for organizations it can be of use to know what causes employees to voice, and under which circumstances the quality of the voice behaviour is high. Secondly, we will contribute to the development of literature concerning voice quantity and voice quality since these subjects have not yet been extensively investigated.
The following research question will be investigated in this paper: What is the influence of transformational leadership on the relationship between behavioural activation and voice quantity? How does creativity affect the relationship between voice quantity and voice quality? And is the relationship between behavioural activation and voice quality mediated by voice quantity?
Theoretical framework Voice behaviour
Voice behaviour can be seen as extra-role behaviour; the kind of behaviour that is discretionary, goes beyond role expectation, and is beneficial for the organization (Liu et al., 2010). Liu et al. (2010) differentiated three characteristics of voice behaviour. Firstly, voice behaviour is discretionary. This beholds that voice behaviour is not obligated in a sense that it is outlined in a job description, or required by an organization. Secondly voice behaviour is
challenge-oriented. The meaning of the performed behaviour is intended to make positive changes to the status quo of an organization. At last, voice behaviour can be risky in a sense that it could be associated with discomfort. One could be viewed negative upon when stating their
ideas out loud. Secondly it could be harmful for professional relationships and result in destructing social capital (Liu et al., 2010).
When voice behaviour is exhibited, employees are willing to share ideas, information, or concerns that are relevant for the organization. It is not just purely speaking out one’s thoughts; voice behaviour is concerned with improvements, work or organizational-related problems, or a vision that differs from the vision of others. The goal of voice behaviour is to make changes in the current situation, which will lead to organizational improvements (Morrison, 2011). Voice behaviour can be explained by the following example: Karen, who works as a waitress in a restaurant, notices that customers are experiencing long waiting times due to disorganization among waiters. From previous work experience she knows that by implementing a more
structured system among the waiters a huge decline in waiting time will result for the customers. In the next staff meeting she decides to share her view in order to improve the working manners of waiters. She expects that her suggestion will lead to a decline in waiting for customers. The moment she shares her view with her colleagues, she is showing voice behaviour.
In the above example, voice behaviour is directed towards a supervisor. Voice behaviour can also be directed at colleagues. Liu et al. (2010) have conceptualized two types of voice behaviour in this context. First, employees could voice to their supervisor, a process which is called speaking up. Secondly they could also voice their thoughts to their colleagues, a process called speaking out.
In the case of speaking out, a peer is risking its team’s psychological safety. Edmonson (2003, p. 6) defines safety as, ‘as a shared belief that the team is safe for interpersonal risk taking’. If team psychological safety arises, the team will be more salient than its individual members. The members have to hold similar perceptions, the climate is characterized by mutual
respect, and there is a sense of interpersonal trust. Another important influence on team
psychological safety is team leader coaching. When a leader is supportive, coaching, and reacts to questions and challenges in a responsive way, members are likely to conclude that the team is working in a safe environment (Edmonson, 2003; Morrison, 2011). We expect that a safe
environment created by a supporting leader will lead to employees voicing more often.
Contextual influences seem to have a significant influence on whether or not to voice. As stated above, the right leadership style can create environments in which people feel safe to voice their behaviour. For this reason we have included leadership in our study. The concept of
leadership will be discussed in another part of this study. Voice quantity and voice quality
We have conceptualized characteristics and targets of voice behaviour. We now will elaborate more on the action (voice quantity) and content (voice quality) of voice behaviour. Recently a distinction has been made between voice quantity and voice quality (Wolsink, forthcoming).
Voice quantity is about the action itself, so how many times do people speak up. This is related to what actually stimulates people to enact voice behaviour. Morrison (2001) created a framework that divided voice behaviour into three types. The construct voice quantity we use in this survey is composed out of these three types. The first one is suggestion-focused voice, which are voice thoughts about improvements for the organization. The second one is problem-focused voice, when expressing these thoughts an employee expresses their concerns about incidents, practices, or behaviours that he or she interprets as harmful to the organization. The last one is opinion-focused voice; this type of voice beholds views on work-related issues that are not aligned with the perceptions held by others.
Voice quality is about the outcome of voice behaviour. It refers to voice behaviour being ‘useful, changing, improving, dissenting and original’ (Wolsink, forthcoming, p. 7). In this study the construct has been composed out of three elements. The first one is originality, which
questions how unique the idea is. The second, flexibility, is about how relevant the idea is to multiple situations. The third, utility, questions how useful the idea actually is.
In this research we focus on speaking out. The focus lies on employees because we expect that when an employee voices his/hers thoughts towards colleagues, those colleagues, will supply his/hears thoughts with feedback. By sharing their ideas with their colleagues, it is most likely that feedback will be given. If the person processes the given feedback he or she is likely to create ideas of higher quality. Kluger and DeNisi (1996) showed in their study that the effect of feedback on performance showed increased performance.
Besides gaining quality through feedback, one could also state that quality can be enhanced by a high amount of voice quantity (Thompson, 2003). Thompson (2003) argued that by producing many ideas the chance of having a high quality idea is bigger than when few ideas are produced. He also argued that a learning effect occurs when people are producing many ideas. Adler & Clark (1991) explain this learning effect in terms of generating ideas and experience. They state that learning is based on repetition and that this will lead to increasing knowledge and abilities. As experience rises, it should enhance performance. Hubert (1991) describes this as the experience-based learning curve. He states that a positive effect can be found of experience on performance. He describes a research in which performance increases as experience grows. Applying this theory on voice behaviour, it is expected that expertise in creating useful, new, and original ideas will grow if someone practices often.
H1: There is a positive relationship between voice quantity and voice quality.
It has been hypothesized that by learning and processing feedback a person could become more sufficient in creating high quality voice. We also focus here on personal characteristics as they can be of influence on voice behaviour (Detert and Burris, 2007). In order for someone to produce original, new, and useful ideas, a person has to possess the ability to actually come up with these ideas. We therefore focus on the concept of creativity in this research.
The moderating effect of creativity
Amabile (1983) stated that for a product or response to be creative it has to be novel, appropriate for the situation, useful and valuable. When given the ability of creativity, a person should be able to use this ability to generate voice behaviour of high quality. Guilford (1957) distinguished three creative abilities: originality, flexibility, and fluency. Originality refers to ideas that are atypical, and irregular to think of. Ideas that are original approach problems in a new way, quite different from the standard solution the problem used to be solved with. When an idea is applicable in multiple categories, this can be referred to as flexibility. The third ability, fluency, is about the amount of unique ideas or problem solutions a person can generate (Baas, de Dreu & Nijstad, 2008; De Dreu, Nijstad, Baas, Wolsink & Roskes, 2012).
The study of Wolsink (forthcoming) suggests that creativity and voice quality are related to each other. She found that higher attention control leads to more voice quality. Attention control leads to systematically analysing combinations of possibilities. By doing so chances are that combinations are produced which can be turned into ideas of high quality. Higher attention control and creativity are related to each other in two ways. First, higher attention control keeps novel ideas activated and discriminates between relevant and irrelevant information. Secondly, creativity activates tendencies such as the production of common ideas and turning these into
new ideas which are useful for solving problems (De Dreu et al., 2012). As higher attention control and creativity are quite similar to each other we expect creativity to lead to high voice quality.
Based on the previous hypothesis which stated that high voice quantity will lead to high voice quality, we predict that this relationship is positively moderated by creativity. A person who has the ability to generate new and useful ideas, is likely to ‘see’ a correct solution for a perceived problem, and therefore able to produce ideas that are useful and will improve the current situation (De Dreu et al., 2012). When one is voicing often, it is likely that through learning (Adler & Clark, 1991) the quality of voice also gets higher. The addition of having a sense for creativity would mean that this process will probably be accelerated as that person will become more competent in generating useful, novel, and appropriate ideas. Out of this the next hypotheses are stated:
H2: There is a positive relationship between creativity and voice quality. H3: The relationship between voice quantity and voice quality will be positively moderated by creativity.
Behavioural activation and voice quantity
Gray developed a theory about brain functions and behaviour, which consisted of two dimensions of personality: impulsivity, which regulates appetitive motivation and anxiety, which regulates aversive motivation (Carver & White, 1994). This paper only focuses on the impulsive dimensions of Gray’s theory. We focus on the impulsive dimensions because we want to
examine what causes people to engage in voice behaviour. We expect impulsivity one of the reasons people exert more voice behaviour.
The behavioural activation system (BAS), it the system that controls appetitive
behaviour, and is sensitive to signals of reward, nonpunishment, and escape of punishment. Once this system is activated, a person starts moving toward a goal. When the BAS system is
activated, feelings of happiness, hope, and elation are evoked. A person with behavioural activation will engage in more goal-directed behaviour and will experience positive feelings when being exposed to impending rewards (Carver & White, 1994).
De Dreu et al. (2010) showed that behavioural activation is positively related to
creativity. In their study they introduce the Dual Pathway to Creativity Model. This model states that any trait or state which increases the amount of cognitive activity or persistence in
generating ideas can lead to creative performance. They provide three reasons for why behavioural activation is a driver of creative performance. First, behavioural activation is correlated with extraversion, and extraversion is correlated to creativity. Second, behavioural activation is associated with moods that can activate cognitive flexibility and creativity. It seems to be that when one is joyful or happy, creativity increases. The third reason states that
behavioural activation is associated with global processing of information, rather than local processing. This means that there is access to a more general amount of information than to a specific kind of information.
The information given leads us to the conclusion that behavioural activation leads to more cognitive activity and persistence in generating ideas (De Dreu et al., 2010). A person with behavioural activation is eager to work towards fulfilling goals, so it is presumable that this person will try to voice ideas as much as possible (Carver & White, 1994). This can be further explained with the expectancy theory of work motivation (Ilgen, Fisher & Taylor, 1979). This theory states that persons create cognitions about how much their behaviour complies with
acquiring a certain reward. The stronger these created associations, the more that persons believes that the attainment of the rewards depends on his/her behaviour. We state that when a behavioral activated person thinks that he/she can attain a rewards by showing voice behaviour that persons will voice more often.
Thirdly, persons with behavioural activation do not take punishment into account, so they are not afraid of the psychological risk they take when voicing. The fourth hypothesis is
formulated as followed:
H4: There is a positive relationship between behavioural activation and voice quantity. The moderating effect of transformational leadership
As has been stated by Detert and Burris (2007), voice behaviour can also be influenced by organizational context. We focus on transformational leadership as these leaders are known for having the ability to develop, stimulate their followers intellectually, and inspire their
followers to set aside their self-interest and reach for a higher collective purpose, while focusing on longer term goals (MacCann, Rawford, Rawlings, 2006). Such a leader also places emphasis to inspire its followers, and motivate them to self-development (Bass, 1999). Transformational leaders have an interest in changing the existing order of things (Conger, 1999). This means that they might stimulate employees in a way that they will engage in voice behaviour and thereby contribute positively to the organizations.
The concept of transformational leadership has been divided by Bass (cited in Den Hartog, 2001, p.176) into four dimensions. ‘Idealized Influence’ (Charisma) is the first dimension, which is needed for a leader to provide a vision and a mission. Charisma can be a used as a method to gain trust and respect. It also leads to a high commitment of followers (Bass & Steidlmeier, 1999).
The second dimension is ‘Inspirational Motivation’ by which leaders let their followers imagine an attractive future. This is done by optimistic enthusiastic talks and encouragement (Bass, 1997; Bass & Steidlmeier, 1999).
Intellectual Stimulation, the third dimension, is a process in which the leader revises the old assumptions, tradition and beliefs. By revising they let their followers think critically about the old days, and thereby stimulate them to see the world in new perspectives. Followers are being stimulated to express their ideas (Bass & Steidlmeier, 1999; Bass, 1997).
The fourth dimension, ‘Individualized Consideration’, entails that leaders treat their followers as individuals. This means that they acknowledge their individual needs, abilities and aspirations (Bass & Steidlmeier, 1999; Bass, 1997).
The last two dimensions, intellectual stimulation and individualized consideration, are aimed at stimulating employees to express their ideas and their aspirations. This could result in a higher amount of voice quantity. Because of the individual focus of these two factors we will group these two dimensions into Individual leadership. The reason we exclude the first two dimensions, idealized influence and inspirational motivation, from this survey is that these dimensions mainly focus on vision and mission, and thereby probably have no influence on stimulating employees in engaging in voice behaviour. These two dimensions are grouped into Inspirational leadership.
As stated earlier, contextual influences seem to have a significant influence on whether to voice or not. In this research we focus on voice behaviour towards colleagues. It has been shown that when a leader is supportive, coaching, and responses to questions and challenges in a
responsive way, members are likely to conclude that the team is working in a safe environment (Edmonson, 2003; Morrison, 2011). When a safe environment is present, people are more likely
to voice their thoughts. It can be presumed that, when a transformational leader is present who encourages his employees to come up with ideas and initiatives, employees will show more voice behaviour.
Liu et al. (2010) showed that the presence of a transformational leader facilitates speaking out. Furthermore, transformational leaders want their followers to contribute in a benefiting manner to the organization. A person who voices will most likely be rewarded for their behaviour. This makes engaging in voice behaviour attractive for a person with behavioural activation, as that person is eager for rewards. Rewards in this manner could be the
acknowledgment of the leader, or obtaining a goal. The next hypotheses are based on the prior arguments:
H5: There is a positive relationship between transformational leadership and voice quantity.
H6: The relationship between behavioural activation and voice quantity is positively moderated by transformational leadership.
As two models have been hypothesized into four hypotheses, this study also aims at testing one model in which the whole model is tested. Carver and White (1994) state that a person with behavioural activation is likely to exhibit ideas, thoughts, and behaviour in order to gain rewards or successes. We have already hypothesized that this behaviour is most likely to lead to a high amount of voice quantity (De Dreu et al., 2010). Also, we have hypothesized that voice quantity, is positively linked to voice quality. According to the model, this means that a person with behavioural activation is likely to voice ideas of high quality, and that this process is mediated by voice quantity. The fifth hypotheses of this study:
H7: A person with behavioural activation will voice high quality ideas towards colleagues and this is mediated by voice quantity.
Design
Two dependent variables will be tested in this research namely, voice quantity, and voice quality.
First it will be tested if voice quantity predicts voice quality. We expect that voice quantity has a positive effect on voice quality. Secondly, it will be tested if creativity has a positive influence on voice quality. Thirdly, it will be tested if creativity has a moderating influence on the relationship between voice quantity and voice quality. We expect voice quality to be highest when creativity is high and when voice quantity is high.
In another model it will be tested if behavioural activation predicts voice quantity. We expect a positive influence of behavioural activation on voice quantity. Secondly we will test if transformational leadership has a positive influence on voice quantity. Thirdly it will be tested whether transformational leadership has a moderating influence on the relationship between behavioural activation and voice quantity. We expect voice quantity to be highest when behavioural activation is high and when transformational leadership is high.
The last design that will be tested is whether voice quantity has a mediating effect on the relationship between behavioural activation and voice quality. Regression analyses will be used for all hypotheses.
Method Sample
A total of 70 triads completed the survey, which leaded to 210 completed surveys. A triad was composed out of one supervisor en two employees who worked together. These participants had to answer questions about one another. Employees had to fill out questionnaires concerning themselves. They also had to rate their colleagues in terms of voice behaviour. All respondent were Dutch-speaking. From the 210 participants, 113 were male (53.81%), whereas 97 were female (46.91%). Table 1 shows demographics about the participants of this study. The organization type the respondents worked in was divided as follows: 75 worked in private companies (35.71%), 36 worked in retail organizations (17.14%), 9 work at health care institutions (4.29%), and 27 (12.86%) work in another type of organization.
Table 1
Summary of demographics of participants
Supervisors Employees Age (years) 42.07 (SD = 11.93) 33.21 (SD = 12.31) Tenure (months) 135.44 (SD = 127.07) 96.40 (SD = 170.32) Male 45 (64.29%) 68 (48.57%) Female 25 (35.71%) 72 (51.43%) Procedure
The data collection was done by seven students of the University of Amsterdam. These students collected the participants by contacting employees or supervisors through email or telephone. This was mostly done by contacting people out of personal networks. A reward was
also included, one supervisor could win 60 euro’s, and one employee could win 250 euro’s. All the participants had to be Dutch and work full time. An email was sent to every person of the triad the evening of the appointment. This email contained a reminder of the appointment made. At the meeting the student shortly gave instructions about the questionnaires. Before
participating, all the participants filled out the informed consent. Once the informed consent was filled in the student send an email to the participants containing a link which led to the
questionnaires. The supervisors and employees had to fill in the questionnaire in a quiet room. All participants also received a personal code, which they had to fill in before starting the questionnaire. The questionnaire for the supervisor took about fifteen minutes to complete. The questionnaires for employees took about twenty-five minutes to complete including the creativity test. After the questionnaires were completed the participants received a small present and an email in which they were thanked for participating.
Measurements
This study uses data collected from 4 scales, namely Behavioural Activation System Questionnaire (13 items), Transformational Leadership Questionnaire (12 items), Voice Quantity Questionnaire (15 items), and Voice Quality Questionnaire (28 items). To measure
Creativity, an assignment was used which will be further explained in this section. A summary of the scales is given in Table 2.
The Behavioural Activation System Questionnaire. This scale measured the behavioural
activation of a participant. It was developed by Carver & White (1994). The scale is composed out of 13 items with an answer Likert scale ranging from 1 – 7.
A high score means that a person experiences many positive feelings when signals of rewards are present, and is likely to behave to attain a reward. A low score means that a person is
less likely to experience positive feelings when signals of rewards are present, and thereby is less likely to move towards a reward.
The Transformational Leadership Questionnaire. This scale measured the amount of
transformational leadership an employee attributes to their supervisor. An employee had to rate their supervisor on 12 statements. This scale was developed by De Hoogh, Den Hartog & Koopman (2004), and existed out of 12 items, with an answer Likert scale ranging from 1 – 5.
A high score means that the supervisor is as a highly transformational leader, so he or she raises feelings of motivation and trust among his employees. A low score means that the
supervisor does not possess many characteristics of a transformational leader. This can mean that he or she does not raise feelings of motivation to contribute to a collective goal and trust among its employees.
The Voice Quantity Questionnaire. Employees had to evaluate their co-worker by
responding to 15 statements. This scale was developed by Wolsink (2013), and exists out of 15 items with an answer Likert scale ranging from 1 – 5.
A high score means that the employee rated by the participant is a person who often suggests many ideas and suggestions. This means that the rated co-worker comes up with a lot of new ideas, and frequently speaks out his own vision and ideas. A low score means that the rated co-worker does not come up with a lot of new ideas, and rarely speaks out his own vision and ideas.
Voice Quality Questionnaire. Employees had to responds to 28 items in which they
evaluated their co-workers. This scale was developed by Wolsink (2013), and exists out of 28 items with an answer Likert scale ranging from 1 - 7.
A high score means that the employee who was evaluated produces ideas or suggestions that are seen as very useful or original. A low score means that the ideas or suggestions proposed by the rated colleague are not seen as useful or original.
Creativity. This test measured the creative ability of the participants. Participants had to
write down as many new ideas in how to use a brick. The minimum time they had for this task was four minutes, after this time a short break was inserted. After this they could choose to extend this period with four extra minutes. They had to write down short but powerful sentences on a computer. This test was derived from similar tests used in the study of De Dreu et al. (2012), and Friedman, Fishback, Forster and Werth (2003). The output of the test was judged independently by two persons. Six factors had to be counted or evaluated. The factors that had to be counted were (1) fluency: number of unique ideas generated per participant; (2) cognitive flexibility: number of categories in which ideas could be classified; (3) persistence: number of unique ideas / number of categories from which they were sampled. One factor had to be evaluated, (4) originality: how atypical and irregular an idea was.
Table 2
Summary of scales including factors, example questions and reliability
Scale Factors Example question
BAS (α = .817) Reward responsiveness (5 items) I feel excited and energized if I achieve something that I want. Drive (4 items) I will do everything to achieve to get what I want.
Fun seeking (4 items) I need excitement and new experiences.
Transformational
Idealized Influence (3 items)
My supervisor …- Radiates confidence about his/her vision and ideas.
Analyses and Predictions
Analyses. To analyse the three models in this study we used SPSS. First it will be tested
if voice quantity has a positive effect has a positive effect on voice quality. Also it will be tested if creativity has a positive effect on voice quality. Lastly we test if the relationship between voice quantity and voice quality is moderated by creativity. Secondly it will be tested if behavioural activation has a positive effect on voice quantity. We also test if transformational leadership has a positive effect on voice quantity. Lastly we test if the relationship between behavioural
activation and voice quantity is moderated by transformational leadership. The last model will (α = .946)
Intellectual Stimulation (3 items)
My supervisor …- Encourages employees to new ways of thinking about problems.
Individualized Consideration (3 items)
My supervisor …- Encourages employees to develop their talents.
Voice quantity
Suggestion-focused (5 items)
How often does your colleague come up with…- suggestions for improvement.
(α = .897) Opinion-focused (5 items)
How often does your colleagues come up with …- advice that deviates from the norm.
Problem-focused (5 items)
How often does your colleagues come up with …- pointing out on inefficiencies in the workplace. Voice quality Originality-focused (14 items) If my colleague comes up with an idea, it is original.
(α = .773)
Flexibility-focused (3 items)
If my colleague comes up with an idea he/she has already been thinking about variations on this idea.
test if behavioural activation has a positive effect on voice quality, and if this is mediated by voice quantity.
Predictions. A positive relation between voice quantity and voice quality is expected.
Subsequently, it is expected that creativity enhances the relation between voice quantity and voice quality positively. Voice quality will be highest under the combination of high voice quantity and high creativity.
Furthermore a positive relationship between a person with behavioural activation and voice quantity is expected. Finally it is expected that a transformational leader enhances the relationship between a person with behavioural activation and voice quantity. Voice quantity will be highest under the combination of high behavioural activation and a high transformational leader.
At last a positive relationship is expected between behavioural activation and voice quality. This effect is mediated by voice quantity. Voice quality will be highest when the behavioural activation is high, and when voice quantity is added to the model.
Results Participants
Analysis was done based on data collected amongst 76 triads of which 70 resulted in completed surveys. Due to missing values we deleted six triads.
Exploratory Factor Analyses
For all key variables an Exploratory Factor Analysis (EFA) with oblique rotation (Direct Oblimin) was conducted.
Transformational leadership. The number of factors that SPSS extracted was two (Table
3). We stated in the theoretical framework, that transformational leadership is divided into two constructs (construct 1: intellectual stimulation and individualized consideration; construct 2: idealized influence and inspirational motivation). The two constructs explained 71.18% variance together. Due to theoretical reasons one item of construct 1 was removed. After removing this item the total explained variance was 71.13% from which construct 1 explained 62.67% variance, and construct 2 8.45% variance.
Table 3
Summary of Factor Analysis of Transformational Leadership
Component Intellectual stimulation and individualized consideration Inspirational motivation and idealized influence My supervisor …- Talks to his employees about what is
important to them. .759
My supervisor …- Encourages employees to new ways of
thinking about problems. .68
My supervisor …- Has a clear vision of the future. .875
My supervisor …- Radiates confidence about his/her
vision and ideas. .544
My supervisor …- Encourages employees to think
independently. .962
My supervisor …- Involves employees in decisions that
My supervisor …- Encourages employees to develop their
talents. .832
My supervisor …- Shows the meaning of working
towards an important and common goal. .730
My supervisor …- Shows to believe in his/her ideals,
beliefs and values. .866
My supervisor …- Is continually looking for new
opportunities for the unit or organization. .671
My supervisor…- Is able to excite others for his/her plans.
Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Bold items are deleted.
Voice quantity colleague rated. As was argued in the theoretical framework voice
quantity exists of three components (Table 4). The three components are suggestion-focused voice, problem-focused, and opinion-focused voice. We set up SPSS to extract three components for the factor analysis. These explained a total of 58.64% variance. The division of items was not spread equally. Five factors explained variance in opinion focused-voice, six factors in
suggestion-focused voice, and two factors in problem-focused voice. We decided to reduce the components back to two and merge opinion- and problem-focused voice. Two components remained which explained 50.65% variance. In this setting three items show low communalities. We decided to remove these items. Explained variance rises to 55.89% in this setting of which
45.62% is explained by opinion/problem-focused voice, and 10.26% variance is explained by suggestion-focused voice.
Table 4
Summary of Factor Analysis of Voice quantity colleague rated
Component
Opinion /problem- focused voice
Suggestion-focused voice
How often does your colleagues come up with…- suggestions for improvement.
How often does your colleague come up with …-new ideas.
-.594
-.818 How often does your colleagues come up with …-
spontaneous inspirations. -.746
How often does your colleagues come up with …-
solutions for problems within the organization. -.505
How often does your colleagues come up with …-
his or her opinion, even if it is different from others. .517 How often does your colleagues come up with …-
advice that deviates from the norm. -.734
How often does your colleagues come up with …-
his or her opinion on matters within the company. .850 How often does your colleagues come up with …-
remarks that involve issues within your company. .765
plans to change things.
How often does your colleagues come up with …- an opinion on the current state of affairs within the
company. .775
How often does your colleagues come up with …- his or her concerns about processes that are not
functioning optimally. .766
How often does your colleague come up with
…-pointing out on inefficiencies in the workplace. .623
How often does your colleague come up with …-personal advice.
How often does your colleague come up with …-technical tips.
How often does your colleague come up with …- raising the issue of things that cannot pass muster.
Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Bold items are deleted
Voice quality colleague rated. We have theorized that voice quantity can be separated
into two constructs, originality and usefulness (Table 5). If we apply this in SPSS a total of 50.20% variance is explained by two constructs. After removing the counterbalanced items which show low communalities, the explained variance rises to 58.80% of which 49.86% variance is explained by originality and 8.92% is explained by usefulness. One item showed a
low communality in this setting. We removed this item which causes the explained variance to rise to 60.27% of which 50.90% is explained by originality, and 9.37% is explained by
usefulness.
Table 5
Summary of Factor Analysis of Voice quality rated
Component Originality Usefulness
If my colleague comes up with an idea, it is original. .863
My colleague has a refreshing view. .651
The suggestions of my colleague are original. .853 If my colleague puts forward an idea, this is innovative. .772 My colleague comes with original opinions. .899 The ideas of my colleagues are innovative. .745 If my colleague gives an opinion, this is refreshing. .581
The opinion of my colleague is original. .811
Suggestions that my colleague makes are innovative. .757 If my colleague comes with a suggestion comes, this is usually
innovative. .798
If my colleague comes up with an idea he/she has already been
thinking about variations on this idea. .514
My colleague thinks about the implementation of his/her ideas. .796 If my colleague gives his/her opinion, these behold important
matters. .825
If my colleague tells his/her worries about processes that are
Ideas of my colleagues can be implemented under different
circumstances. .678
If my colleague comes up with a new product/method, this is
worked out accurately. .579
The ideas of my colleagues are well applicable. .788
The opinions of my colleagues are helpful. .805
If another comes up with an idea, my colleague can suggest
useful additions. .649
The opinions of my colleagues are not innovative. Suggestions of my colleagues are predictable.
The opinion of my colleagues does not show me any new insights.
Ideas of my colleagues are predictable.
The ideas of my colleagues can be poorly applied in practice.
If my colleague makes a suggestion, it is most likely useless. My colleague creates ideas for our department, and also for other departments within the company.
Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Bold items are deleted and counterbalanced
BAS. As we have a theoretical base for dividing BAS up into three groups, extraction in
removing one item due to a low communality the total variance explained was 58.46%. Drive explained 34.82%, fun seeking 12.7%, and reward responsiveness 10.93% variance.
Table 6
Summary of Factor Analysis of BAS
Component Drive
Fun seeking
Reward responsiveness I feel excited and energized if I achieve something that I
want. -.844
Winning a match would make me really excited. -.552
If I'm good at something, I put an effort into it. -.352
If positive things happen to me, it has a strong influence
on me. -.737
I get enthusiastic if I see an opportunity to get what I
want. -.711
I will do everything to achieve it to get what I want. .822 When I see a chance to get what I want, I will take that
chance. .559
I go out of line to get what I want. .686 If I want to achieve something, nothing can stop me. .837
I often behave impulsively. .761
I need excitement and new experiences. .362
I often do things just for fun. .767
Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Bold items are deleted
Correlations
To see if the proposed hypotheses are supported, correlations between key variables will be discussed. A correlation indicates to what extend and in what measure these variables are connected to each other (Siero, Huisman, & Kiers, 2009). All correlations can be found in Table 7.
The first correlation concerned the relationship between voice quantity and voice quality. We expected a positive relationship between voice quantity and voice quality, this was
confirmed by the results (r = .522, p < .05). This indicates that someone who voices often, many times voices ideas of high quality. Someone who does not voice often, voices ideas of less quality. In other words, voice quantity seems to relate to voice quality.
The second correlation concerned the relationship between creativity and voice quality. The results, r = .246, p < .05, confirmed our expectation; someone who is creative, voices ideas of high quality, and someone who is less creative, voices ideas of lower quality. This can indicate that creativity relates to high voice quality.
The third correlation concerned the relationship between behavioural activation and voice quantity. No relation was found between these constructs (r = -.069, p > n.s.). There seems to be no relationship between behavioural activation and voice quantity, this means that the two constructs do not seem to be connected to each other.
The fourth correlation showed that there was no relationship between transformational leadership and voice quantity (r = -.046, p = n.s.); there is no connection between
transformational leadership and voice quantity.
The fifth correlation, the relationship between behavioural activation and voice quality, showed a negative relationship, (r = -.233, p < .05). This could indicate that behavioural activation has a negative influence on voice quality. It could mean that someone who has high behavioural activation is more likely to voice ideas of low quality, and someone who has low behavioural activation is more likely to voice ideas of high quality. In other words, behavioural activation seems to be negatively related to voice quality.
To investigate these relationships more deeply, we will test the hypotheses with a regression analyses.
Table 7
Correlations between key variables
BAS Voice quantity Voice quality Creativity Voice quantity -.069
Voice quality -.233** .522**
Creativity -.152 .098 .246**
Transformational leadership .178* -.046 -.053 -.129 *. Correlation is significant at the 0.05 level (2-tailed).
Moderation voice quantity, voice quality and creativity
This model was tested by using a regression analysis conducted by SPSS, of which the results can be found in Table 8. Firstly it was tested whether voice quality was higher when voice was quantity high then if voice quantity was low. Table 8 shows that the effect of voice quantity on voice quality was supported by the results (β = .522, p < 0.05). This supports H1: There is a positive relationship between voice quantity and voice quality. This means that people who show a higher level of voice quantity also show a higher level of voice quality.
Secondly it was tested whether creativity influenced voice quality. We expected that a high level of creativity would lead to higher voice quality. This relationship was founded positive as expected, β = .246, p < .05. This supports H2: There is a positive relationship
between creativity and voice quality. This means that people who are highly creative voice ideas of higher quality.
Thirdly it was tested if there was an interaction effect of voice quantity and creativity on voice quality. It was expected that creativity positively influenced voice quantity when voice quantity was low, but not if voice quantity was already high. The hypothesis (H3), the
relationship between voice quantity and voice quality will be positively moderated by creativity, is accepted (β = .457, p < .05). Figure 1 shows the interaction effect for more clarification. It
β = .522 22 Β = .522 β = .246 22 Β = .522
seems to be that creativity is not needed to voice ideas of high quality when one is voices often. When one is not voicing often creativity is needed in order to voice ideas of high quality.
Table 8
Regression results Voice quantity, creativity and voice quality (dependent variable)
Beta t
Intercept 0
Voice quantity (A) .522** 7.33
Creativity (B) .246** 3.04
A * B .457** 6.17
*. Significant at the 0.05 level (2-tailed). **. Significant at the 0.01 level (2-tailed).
Figure 1. Moderation effect of voice quantity, quality and creativity 1 1,5 2 2,5 3 3,5 4 4,5 5
Low Voice quantity High Voice quantity
De p en d en t var iab le Low Creativity High Creativity
Moderation BAS, voice quantity and transformational leadership
This model was tested by using a regression analysis conducted by SPSS, of which the results can be found in Table 9. Firstly it was tested whether voice quality was higher when behavioural activation was high then if behavioural activation was low. Table 9 shows that the effect of behavioural activation on voice quality was not confirmed by the results (β = -.069, p > n.s.). The hypothesis (H4): there is a positive relationship between a person with a behavioural activated system and voice quantity, is rejected. This means that people with a high level of behavioural activation do not show high levels of voice quantity.
Secondly it was tested whether transformational leadership influences voice quality. We expected that voice quantity was higher when transformational leadership was high then if transformational leadership was low. This relationship was not found (β = -.046, p = n.s.). H5: There is a positive relationship between transformational leadership and voice quantity, is rejected. This indicates that voice quality of employees with a transformational leader was the same for people that do not have a transformational leader.
Thirdly it was tested if there was an interaction effect of behavioural activation and transformational leadership on voice quality. It was expected that transformational leadership positively influenced an employee with high behavioural activation, but not if the employee had
β = -.069 22 Β = .522 β = -.046 22 Β = .522
low behavioural activation. The hypothesis (H6): the relationship between a person with
behavioural activation and voice quantity is positively moderated by transformational leadership is rejected (β = -.075, p = n.s.). This means that there is no difference in voice quantity for either high or low behavioural activation when the leader performs transformational leadership
behaviour.
Table 9
Regression results of BAS, transformational leadership , and voice quantity (dependent variable)
Beta t Intercept 0 BAS (A) -.069 -.814 Transformational leadership (B) -.046 -.539 A * B -.08 -.887
*. Significant at the 0.05 level (2-tailed). **. Significant at the 0.01 level (2-tailed).
Mediation BAS Voice quantity and voice quality
The mediation effect is tested by using regression analysis in SPSS, of which the results can be found in Table 10. To test if H7: A person with behavioural activation will voice high quality ideas towards colleagues, and this is mediated by voice quantity, we first looked at the relationship between behavioural activation and voice quality. Unexpectedly, we found a
negative effect of behavioural activation on voice quality while we expected a positive one (β = -.233, p = < .05). This indicates that people in our sample with high behavioural activation voice lower quality than people with low behavioural activation. Secondly we looked at the
relationship between behavioural activation and voice quantity. We did not find the expected effect (β = -.069, p = n.s.), which means that there is no difference in voice quantity if
behavioural activation is either high or low. The third relationship we investigated was between voice quantity and voice quality which was positive, as expected (β = .522, p < 0.05). This means that people who show a higher level of voice quantity also show a higher level of voice quality. Mediation is not possible because there is no relationship between behavioural activation and voice quantity; H7 has been rejected. This model does show us two main effects; voice quality can be predicted by high voice quantity; and behavioural activation leads to low voice quality. β = -.069 22 Β = .522 β = .-.233 22 Β = .522 β = .522 22 Β = .522
Table 10
Regression results of BAS, voice quantity , and voice quality
(dependent variable)
Model 1 Model 2 Model 3 Model 4
Dependent variable Voice quantity Voice quality Voice quality Voice quality
Beta t Beta t Beta t Beta t
Intercept 0 0 0 0 BAS -.069 -.814 -.233** -2.834 0.504 7.125 Voice quantity .522** 7.33 -0.198 -2.803 R2 0.069 0.233 0.554 0.554 *. Significant at the 0.05 level (2-tailed).
**. Significant at the 0.01 level (2-tailed).
Discussion Summary of the results
The main goal was to test whether behavioural activation and transformational leadership had an influence on voice quantity and whether voice quantity and creativity had an influence on voice quality.
Voice quantity, voice quality and creativity. First we will look at the relationship
between voice quantity and voice quality. The data supported the expected effect: (H1) There is a positive relationship between voice quantity and voice quality. Multiple theories can explain the founded effect. Firstly voice quality can be heightened by the effect of feedback on
on its voice behaviour. This feedback, which is given by colleagues, would give the person suggestions for improvements, which leads to higher voice quality. A second explanation can be clarified with the experience-based learning curve (Hubert, 1991). According to this theory voice quality improves trough a learning effect (Adler & Clark, 1991; Hubert, 1991). As one voices more often experience grows. This leads to increasing knowledge and abilities and will lead to an increase in performance. The result of this process is that voice quality will be higher.
Secondly we will look at the influence of creativity. We expected that a creative person would produce high quality voice. H2: There is a positive relationship between creativity and voice quality is supported. Wolsink (forthcoming) argued that attention control leads to more creativity. We stated that creativity and attentional control are related to each other as attention control keeps novel ideas activated and discriminates between relevant and irrelevant
information. Creativity activates tendencies which turn common ideas into new ideas which are useful for solving problems. This way we concluded that creativity would lead to voice quality. There seem to be two very plausible ways to voice quality; voicing often and being creative.
Thirdly we will look at the effect that (H3), the relationship between voice quantity and voice quality will be positively moderated by creativity. As can been seen voice quality is most high under the combination of high creativity and high voice quantity. Remarkable is that when voice quantity lowers a creative person still seems to produce voice of high quality. The two separate theories do not seem to exclude each other but work in an additive way. Creativity heightens voice quality (De Dreu et al., 2012). Voice quality becomes even higher when voice quantity is also high. In other words, someone who is creative is able to voice high quality ideas and by increasing voice quantity a learning effect occurs which causes this person to voice ideas
of even higher quality (Adler & Clark, 2012; Hubert, 1991). Subsequently if one is not creative but voices often, a learning effects can occur which will lead to higher voice quality.
BAS, voice quantity and transformational leadership. First we will look at the
relationship between behavioural activation and voice quantity. There is no support for our expected effect that (H4) there is a positive relationship between behavioural activation and voice quantity. We argued that behavioural activation would motivate a person to voice often as voicing can lead to a reward. A reward could be that one could be praised when voicing
something that could be of good for the company. It could be that we interpreted this wrongfully. Behavioural activated persons want to achieve something when signals of rewards are present (Carver & White, 1994). Highly behavioural activated persons do not engage in voicing as there is no impending reward. This could be an explanation for not finding a relationship between behavioural activation and voice quantity. This argument also explains why our prediction based on the expectancy theory of work motivations was not supported. We proposed that a
behavioural activated person would show more voice quantity if they believed that they could attain a reward by voicing more often (Ilgen, Fisher & Taylor, 1979). We have already argued that the outcome of voicing is not seen as a impending reward by behavioural activated persons. When there is no impending reward there is no need for a person to create associations based on which behaviour should be performed to attain the reward (Ilgen, Fisher & Taylor, 1979).
We also looked at the influence of leadership in this model. We stated that the presence of a transformational leader was positively related to voice quantity. However, the hypothesis (H5) there is a positive relationship between transformational leadership and voice quantity, is not supported in this study. We did not find an effect of transformational leadership on voice quantity. We expected that Inspirational leadership (individualized consideration and intellectual
stimulation) would predict that a higher amount of voice quantity as they stimulate their employees to express their ideas and their aspirations. Edmonson (2003) and Morrison (2011) showed that a supportive leader let its employees feel safe in their work environment which results in more voice behaviour. Furthermore Liu et al. (2010) showed that a transformational leader facilitates the process of speaking out. We expected that employees would engage in more voice behaviour towards each other with the presence of a transformational leader. The reason we found no effect could be that the influence of transformational leadership manifested itself on the concept of speaking up. It could be that the influence of the leader resulted in a higher
amount of voice quantity directed at the leader itself. Furthermore voice behaviour should not be seen as extra role behaviour which is encouraged by transformational leaders (Bass, 2001). Voice behaviour is meant to contribute positively to an organization, but it could also entail criticism. Therefore voice behaviour is not something that a transformational leader would encourage towards its employees.
Thirdly we expected a transformational leader to positively influence an employee with high behavioural activation, but not an employee with low behavioural activation. The
hypothesis (H6): the relationship between behavioural activation and voice quantity is positively moderated by transformational leadership, is not supported. We expected that a transformational leader would stimulate a behavioural activated person to engage in more voice behaviour. An explanation for not finding this effect is that voice behaviour is not attractive for highly
behavioural activated people. Those people will not engage in more voice behaviour even if it is wanted by a transformational as the outcome of voice behaviour is not interesting for them.
Mediation BAS, voice quantity and voice quality. We expected that behavioural
voice quantity. No mediation effect was found and H7: A person with behavioural activation will voice high quality ideas towards colleagues, and this is mediated by voice quantity is therefore not supported. The model does show us two main effects; voice quality can be predicted by high voice quantity; and high behavioural activation predicts low voice quality. The reason that high voice quantity is an antecedent of voice quality is probably due to the fact that a learning effect occurs (Adler & Clark, 1991). By voicing often, knowledge/expertise increases, which enables a person in this case, to create high quality voice.
Interesting is that persons with high behavioural activation show lower voice quality. Person with high behavioural activation do not engage in voice behaviour. This results in showing less voice quantity. A learning effect will not arise among these people and voice quality will not be higher in future voicing (Adler & Clark, 1991). Also when one voices rarely chances of creating a good idea are low (Thompson, 2003). No relationship was found between behavioural activation and voice quantity. Reasons for this effect are explained above.
Contributions
The most important contribution of this study is that voice quantity increases voice quality. If we take a deeper look into the model by adding creativity to it, a positive relationship seems to exist. These effects can mean that there are two ways to increase voice quality. First, high voice quality can be achieved by being creative. This complies with the findings of De Dreu et al. (2012) who states that creativity activates the generation of producing ideas, and adjusting these ideas so that they are applicable to problems. Also this complies with the study of Detert and Burris (2007), which stated that personal characteristics can influence voice behaviour of employees. Secondly voice quality can be achieved by voicing often. This study shows that a high amount of voicing leads to high voice quality. We proposed that through a learning effect,
voice quality rises (Adler & Clark, 1991; Hubert, 1991). We hereby contribute to the literature by showing that the antecedents of voice quality may be twofold.
Strengths of the study
There were multiple strengths in this study. By delegating the data collection to multiple persons a high amount of data was collected. As everyone contacted their own personal network a variety of people was contacted. This has led to a good distribution of age, gender and
industries. By visiting the participants at their workplaces a high response rate was achieved. Furthermore, visiting the participants has led to reliable data. The study was explained very precise, and the participants completed the study in the same room as the researchers. Implications
Outcomes of this study are intended for companies who benefit by idea generation. For those companies it would be of matter to know what causes employees to voice, and what causes voice quality. We found that ideas of high quality exist when a person is creative and/or voices often. Companies should focus on creating an environment where employees feel safe to voice often. Additionally feedback should be given to voice. This will increase the learning process which could results in higher voice quality.
This study also has implications for literature concerning voice behaviour. Firstly we propose that there are two ways to voice quality; voicing often and creativity. Secondly we have found that behavioural activation does not lead to more voice quantity. It should be considered by researchers if behavioural activation is a reliable predictor of voice quantity/quality as there in no real reward attainable. Thirdly we have found that transformational leadership is a construct that is might better to split up in multiple constructs. Analyses have shown that by splitting the construct into combinations, transformational leadership can be a more reliable predictor.
Limitations and future research
A limitation of this study concerns the measurement of multiple constructs. We measured behavioural activation of which the reliability was high. Considered afterwards, we might have tried to measure this latent variable in a wrong way. Behavioural activation is a controlling mechanism in the brain and thereby difficult to measure in a subjective way. Besides this fact, we already stated above that behavioural activation might not be a good predictor of voice quantity/quality. Future researchers should consider to not include this construct anymore in new studies about voice behaviour. If they do so, behavioural activation should be measured in a more objective manner.
A suggestion of future research might lie in the area of predicting voice quality. We proposed that there might be two paths that lead to voice quality; people either have to be creative, or have to voice a lot. That last aspect could be of matter for future research. As we stated there might an influencing factor of learning in producing voice of high quality. A longitudinal research in which people have to voice ideas could measure if a learning effect really leads to high quality voice. Voice quality could be measured every couple of months. The expectation would be that after every measurement moment the quality of voice is higher.
The last suggestion for future research entails investigation in what way high creative people would voice often. It seems that the higher the voice quantity, the higher the voice quality. Also creative people produce high voice quality. Higher voice quality could exist if the group of highly creative people would voice more often. Researchers could investigate under which circumstances voice behaviour is likely to be exhibited. It could be investigated if
motivation and psychological safety are of influence. By measuring the voice quantity of creative people in a longitudinal setting but in different conditions (high vs. low motivation x high vs.
low psychological safety) outcomes can show us under when creative people are more likely to voice and if there voice quality actually increases.
Conclusion
This study was focused on the antecedents of voice quantity and voice quality. We propose that there might be two ways in order to create voice quality. First, high voice quality can be achieved by highly creative people. Secondly voice quality can be achieved by voicing often. More research should be done in order to determine antecedents of voice quantity among highly creative people. Organizational success can be gained if employees dare to voice their ideas and are given feedback to increase the quality of these ideas.
References
Adler, P. S., & Clark, K. B. (1991). Behind the learning curve: A sketch of the learning process. Management Science, 37(3), 267-281.
Amabile, T. M., & Amabile, T. M. (1983). The social psychology of creativity Springer-Verlag New York.
Baas, M., De Dreu, C. K., & Nijstad, B. A. (2008). A meta-analysis of 25 years of mood-creativity research: Hedonic tone, activation, or regulatory focus? Psychological Bulletin, 134(6), 779.
Bass, B. M. (1999). Two decades of research and development in transformational leadership. European Journal of Work and Organizational Psychology, 8(1), 9-32. Bass, B. M., & Steidlmeier, P. (1999). Ethics, character, and authentic transformational
leadership behavior. The Leadership Quarterly, 10(2), 181-217. Burns, J. (1978). Leadership, 1978. New Yorker: Harper & Row,
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS scales. Journal of
Personality and Social Psychology, 67(2), 319.
Conger, J. A. (1999). Charismatic and transformational leadership in organizations: An insider's perspective on these developing streams of research. The Leadership Quarterly, 10(2), 145-179.
Conger, J. A., Kanungo, R. N., & Menon, S. T. (2000). Charismatic leadership and follower effects. Journal of Organizational Behavior, 21(7), 747-767.
Crant, J. M., Kim, T., & Wang, J. (2011). Dispositional antecedents of demonstration and usefulness of voice behavior. Journal of Business and Psychology, 26(3), 285-297.
D'Aprix, R., & Fagan-Smith, B. (2011). The case for an open communication
culture: Companies with high levels of engagement show a return on assets six times higher than companies with low engagement levels. Retrieved 04/09, 2014,
from http://www.industryweek.com/articles/the_case_for_an_open_communication_culture _24357.aspx
De Dreu, C. K., Nijstad, B. A., Baas, M., Wolsink, I., & Roskes, M. (2012). Working memory benefits creative insight, musical improvisation, and original ideation through maintained task-focused attention. Personality & Social Psychology Bulletin, 38(5), 656-669.
doi:10.1177/0146167211435795; 10.1177/0146167211435795
De Hoogh, A., Den Hartog, D., & Koopman, P. (2004). De ontwikkeling van de CLIO: Een vragenlijst voor charismatisch leiderschap in organisaties. Gedrag En Organisatie, 17(5), 354-381.
Den Hartog, D. (2001). Leadership in organizations. Handbook of Industrial, Work & Organizational Psychology: Volume 2: Organizational Psychology, 166.
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.
Dyne, L. V., Ang, S., & Botero, I. C. (2003). Conceptualizing employee silence and employee voice as multidimensional constructs*. Journal of Management Studies, 40(6), 1359-1392. Edmondson, A. C. (2003). Speaking up in the operating room: How team leaders promote
learning in interdisciplinary action teams. Journal of Management Studies, 40(6), 1419-1452.
