Master Thesis
Voicing at the workfloor
~“it is just not possible any longer to 'figure it out' from the top"~ Senge (1990)
Course: Master Thesis
Student: Romy Berends (10009841)
Lecturer: Inge Wolsink
Deadline: June 25th of 2014
Word count: 8488
Abstract
It is no longer possible for managers to figure it out by themselves, they need their employees to voice. Voice is the communication of ideas, suggestions and opinions for constructive change. Previous work considers that voice can be split into two components: Voice Quantity and Voice Quality, and that this could be implemented by linking the components to the dual process theory. The dual process theory explains the two processing modes (Automatic and Systematic) of individuals. We proposed that the automatic system will facilitate Voice Quantity and the systematic system will facilitate Voice Quality. The propositions are tested with dyads of 141 employees and their manager. The proposition that the systematic system facilitates Voice Quality and not Voice Quantity could be confirmed by the results. The proposition that the automatic system only facilitates Voice Quantity could not be confirmed due to methodological issues. Still, the empirical evidence for a systematic processing route to Voice Quality is strong and future research is suggested for the automatic processing route to Voice Quantity.
Content Abstract 2 Preface 5 1. Introduction 6 1.1 Voice Behavior 8 1.2 Attention 11
1.3 Voice Quantity versus Voice Quality against Attention 12
1.4 Impulsivity and Voice Quantity vs Voice Quality 14
1.5 The interaction effect of Attention and Impulsivity 16
2. Method 19
2.1 Design and sample 19
2.2 Procedure, test and measurements 20
2.2.1 Dependent variables: Voice Quantity & Voice Quality 20
2.2.2 Independent variable: Attention 21
2.2.3 Independent variable: Impulsivity 22
2.2.4 Control variables 23
2.3 Analysis and predictions 23
3. Measurements Models 25
3.1 Moderation model: Voice Quantity 25
3.2 Moderation model: Voice Quality 25
4. Results 26
4.1 Descriptive 26
4.1.1 Factor and reliability analysis 26
4.1.2 Correlations & control variables 27
4.2 Regression analysis 28
4.2.1 Regressions and moderation model: Voice Quantity 28 4.2.2 Regressions and moderation model: Voice Quality 30
4.3 Processing modes graphic against Voice Quantity & Voice Quality 32
5. Discussion 33
5.1 Link between theory, hypotheses and results 33
5.2 Influences of the design of the study and other discussion points 37
5.3 Conclusion & contributions 38
References 40
Appendix A: Survey 46
Appendix B: Informed Consent 49
Appendix C: Factor analysis 50
Preface
This thesis is written for my Business studies Master degree at the University of Amsterdam. I wanted to specialize myself during my masters in Human Resource Management and therefore I followed the Leadership and Management track. My master thesis is the last step I have to take before finishing four years of hard work and fun. Sometimes it has been tough, but most of all I have learned a great deal and had fun by doing so. While writing this thesis I had help from several people who I would like to thank. First of all, special thanks to Inge Wolsink, who’s feedback sometimes leaded to major frustrations, but in the end always helped me to improve and to get all the information straight. I can honestly say that I have learned a lot of her and she encouraged me to develop my own competences further. Furthermore, I want to thank everyone who was willing to read my thesis and check in on mistakes and clarity. And last, thanks to all the respondents who were willing to participate in this study.
I hope you will enjoy reading my thesis,
Romy Berends
1. Introduction
~“it is just not possible any longer to 'figure it out' from the top"~ Senge (1990)
Business starts with people. In other words a company cannot exist without people starting and leading it. Of course many parts of the business are automatized, such as digital documenting which orders and saves documents in the right way without the necessity of human input. Only this is not possible for all parts. Especially not for the parts of business where for example the strategies are set out. Setting and implementing a strategy takes more than one person. Teamwork and good communication between business units is necessary; first to decide for a creative and unique strategy and second to implement this strategy along the different business units. The reasons why a specific strategy is chosen and adopted by the other team members remain unclear.
It is no longer possible for managers to figure it out by themselves. Managers face an improving dynamic and fast changing work environment and they need the ideas from their employees and coworkers to stay ahead (Morrison & Milliken, 2000). This study suggests that voice (the communication of ideas, suggestions and opinions for constructive change) has an important role in this strategy process (Wolsink, Forthcoming). Employees can voice towards their colleagues (speaking up) and towards their supervisor (speaking out) (Liu, Zhu &Yang, 2010). It seems that speaking up and speaking out are just as important, but it is the supervisor who is in the position to make a change (Detert & Burrus, 2007). Thus, if employees dare to speak out their ideas to the (line) supervisor this can be more useful than when
communicating it to their colleagues. Furthermore, voice can bring about change when ideas suggest organizational improvements in the status quo that are new, original and agreed by the
supervisor and team members (Van Dyne, Cummings & Parks, 1995). In other words,
suggestions for a change in the known processes and structure of a company (Van Dyne et al., 1995).
Although people agree that voicing at the work floor is important, it is not said that everything voiced contains something useful. Although voicing is driven by the goal to make an improvement, to really say something useful an employee needs a good idea or suggestion (Wolsink, Forthcoming). An idea is good when it is new, original and useful which are the same evaluators as with creativity (Amabille, 1983; De Dreu et al., 2010). If an idea is not original it is merely an repetition from something that already exists. Voice has the goal to improve the status quo into something better, so the originality of the idea voiced seems a necessity. Additionally, if the idea is not useful there is no reason to implement it (Wolsink, Forthcoming). Thus, originality and usefulness are indeed important characteristics of the ideas voiced. Therefore the creativity of the employee is suggested to influence the quality of the ideas voiced. We propose that creativity plays an important role in relation to Voice Quality.
Additionally, the work of De Dreu, Nijstad, Baas, Wolsink en Roskes (2012) suggests that creativity is attributed by systematic thinking. This is the ability to thoroughly explore and elaborate on ideas. For this ability it is beneficial to be able to control conscious Attention for long periods of time. One thus needs the systematic mental capacity to keep conscious Attention focused on a creative and qualitative voice task (De Dreu et al., 2012). Attention distinguishes non-conscious- from conscious thinking (Kahneman & Frederick, 2002). This is described in dual process theories which describe that everyone has two cognitive systems which separate two mental processes: the first is an automatic and the second a systematic process (Evans, 2007). Conscious thinking is systematic and slow (Evans, 2007).
conscious thinking on the other hand emerges in the cognitive mental system that is automatic and rapid (Evans, 2007).
In order to contribute to previous research the goal of this study is twofold. First, this study will focus on both cognitive systems and their different influences on voicing. This to clarify the importance of splitting voice into two components: ‘having to say something’ (quantity) vs ‘saying something useful’ (quality). Former research to the role of automatic and systematic processing routes in relation to voice behavior is rare, and especially research to both routes at the same time has not been done before (Detert & Burris, 2007; Morrison & Milliken, 2000; Morrison, 2011). Second, the practical goal is that we will try to describe the cognitive characteristics the perfect employee should have to voice the most ideas that are high in quality.
1.1 Voice Behavior
This study defines Voice as: an act or verbal expression, where a message is conveyed from an employee to a supervisor (Morrison, 2011). Furthermore, voice behavior is not just communication in or about the organization, alternatively it is self-initiated, change- and improvement focused and involves communicating new ideas for current affairs that could benefit the organization in the future (Wolsink, forthcoming). Additionally, the self-initiated component of voice is much in line with proactive behavior. A proactive person is someone who scans the environment for opportunities, shows initiative and takes action (Bateman & Crant, 1993). Voice thus is a proactive behavior.
For supervisors it is critical to the well-being of their organization that employees speak out. Organizations rely more and more on the ideas of their employees which can be the basis of innovation and quick responses to survive the dynamic environment they are in (Edmondson, 1999; Morrison & Phelps, 1999). Unfortunately this ideas are insufficiently
provided by the employees and still much more an act of their supervisors (Morrison & Phelps, 1990).Therefore it is important to understand who and why a certain person decides to speak up or not.
According to Morrison (2011) there should be some nuances made in the voice literature. She suggests three reasons for employees to voice: suggestion focused voice, providing ideas and suggestions for improvement; problem focused voice, protecting for potential hazards to the organization and opinion focused voice, communicating viewpoints that differ from others. This differentiation raises the importance of a deeper understanding of voice behavior even more. These three types of voice focus on what influences the frequency instead of the quality of voicing.
Additionally, earlier studies show that there are three lines of research focusing on this gap in the understanding of voice behavior. The first line focusses on the individual
differences and demographic characteristics of the persons voicing (Crant, 2003; Van Dyne & Le Pine, 2001). The second line focusses on the loyalty and the feeling of responsibility of the employees (Hirschman, 1970) and the last line of research focusses on factors of the
organizational context that influence the willingness of employees to speak up (Edmondson, 2003; Milliken et al., 2003).
This study will focus on the first line by investigating the two cognitive systems in combination to voice behavior and how this differs for different employees (Evans, 2007). For example; it is suggested that the decision to voice is a rapid and automatic process and thus lacks Attention. This means that the decision to voice is non-conscious. In contrast, to voice useful and original ideas it is assumed that Attention is needed. Attention allows for
systematic processing which is important to be creative and thus to voice ideas of quality. The quality of voicing is thus determined by conscious thinking (Evans, 2007; Wolsink,
Forthcoming).
The last line will also receive focus because the study of Liu, Zhu and Yang (2010) suggests that employees choose to whom they are willing to speak up. In other words, there is a distinction in the targets to whom employees voice: their colleagues or their supervisors. Especially the voice behavior towards supervisors is of importance as they are in the position to act to a critical suggestion (Detert & Burris, 2007).
Previous research based voice behavior of employees mostly on the quantity of the ideas voiced. However, Grant and Ashford (2008) have argued that voice is a dynamic process and only studying the limitations to the voice quantity of voice is insufficient to fully explain voice. A distinction can be made between the frequency and the content of the voice behavior of employees (Wolsink, Forthcoming). We will refer to these differences with Voice Quantity and Voice Quality. With Voice Quantity meaning having to say something and with Voice Quality meaning to say something useful (Wolsink, Forthcoming, p. 3). Employees who have something to say might be driven by the desire to make a constructive change but to say something useful employees require a good idea. An idea is good when it is new, original and useful which are the same evaluators as with creativity (Amabille, 1983; De Dreu et al., 2010). We have argued that creativity and thus Voice Quality requires Attention which is facilitated by the systematic processing mode. This suggests that the content of the idea is influenced by the systematic processing mode of the employee. On the other hand, Voice Quantity is based on the decision to voice. Decision making is argued to be based on emotions and is often fast (Epstein, 1994; Hassin et al., 2005), which suggest that the decision to voice is facilitated by the automatic cognitive system of an employee. Therefore, this study splits Voice into two components to be able to show that they are influenced by two different mental processes.
Why is it important to understand what influences the content of the employees voicing? Currently managers work in a fast changing environment, which means they have to
spent their time efficiently (Morrison & Milliken, 2000). New ideas facilitate continuous improvement (Nemeth & Staw, 1989). It would be a loss of efficiency when managers get lots of quantity but with little quality. For managers it is thus necessary that the ideas voiced are of quality. However, the quality of voiced ideas has not received much Attention in extant literature while this study suggests that it is useful for managers to have employees in their organization who dare to speak up and furthermore have a high amount of quality in their ideas voiced (Ashford & Black, 1996). The former thus leads to the question: What is the role of the two different processing modes (automatic & systematic) and thus Attention in relation to the frequency of idea communication (Voice Quantity) versus the content of idea
communication (Voice Quality)?
1.2 Attention
Attention is a systematic process which makes creativity in the idea generation process
possible and is central to the working memory capacity (WMC) (Kane & Engle, 2003). WMC is the capacity to selectively maintain and manipulate goal-relevant information without getting distracted by irrelevant information over short intervals (Jha, Stanley, Kiyonaga, Wong & Gelfand, 2010). Individual differences in WMC affect the capability to engage in controlled processing and so determines the ability to control thoughts, feeling and action in the course of everyday life. Operationally this means the amount of items that can be recalled in a complex WMC test (Barrett, Tugade & Engle, 2004). The system is essential for keeping task goals and the information relevant to these goals in an active mental state (Harrison, Shipstead, Hicks, Hambrick & Redick, 2013). In other words, the WMC of an employee determines the capability to detect the parts of their environment that need problem solving without getting distracted, if this is relevant to their goals (Unsworth & Engle, 2007). It is therefore assumed that WMC directs the focus of Attention and controls conscious
processing, which allows for systematic processing (Wolsink, Forthcoming). We suggest that an employee’s WMC is an indicator of the ability to control Attention and is stimulated by the systematic processing mode of employees.
1.3 Voice Quantity versus Voice Quality against Attention
Two components of the voice process are distinguished. First, employees have an idea and second employees decide if they are going to voice it. There are several cognitive factors that determine why an employee chooses to voice. Examples are: responsibility, self- protective behavior and loyalty (Morrison & Phelps, 1999; Van Dyne et al., 2003).
The self-protective component explains why employees rather choose to take a passive strategy. The main reason to voice according to this study is to change the status quo. This makes voicing socially risky, employees and people in general are risk aversive (Detert & Burris, 2007). They rather ‘protect’ themselves against the opinion that can follow from their supervisor and colleagues on their suggestion. This view suggests that the choice process, involves a conscious and thus systematic risk calculation. Employees consciously consider if their suggestion will be effective and is furthermore wanted (Lepine & Van Dyne, 1998). Contrasting, Edmondson (2009) proposes that the decision to remain silent (intentionally withholding information and ideas with relevance to the organization) is because of fear and thus based on emotions. Contrasting emotions like happiness can result in the opposite which is speaking out.
The decision to voice influences the frequency of the ideas voiced and thus the Voice Quantity is affected by emotions and risk taking tendencies (Detert & Burris, 2007;
Edmondson, 2009). From the research of Evans (2007) and Epstein (1994) we furthermore derived that decision making is an rapid and automatic process, instead of slow and
systematic like Attention. Hence, the role of automatic versus systematic processing is not
clear in the current literature (Wolsink, Forthcoming). We propose this is because when scholars talk about voice, they fail to see the differences between the decision process (automatic processing) and the effectiveness (systematic idea generation) of the voice
behavior. For these reasons we propose that conscious Attention and keeping a large quantity of information active in memory (WMC), might be of limited influence in the decision
making part of the voice process. Therefore we propose that Voice Quantity is less influenced by Attention than Voice Quality is.
While Voice Quantity is merely influenced by script and emotions, Voice Quality is based on idea generation. Creativity is the generation of ideas, insights or problem solutions that are new yet appropriate (Amabile, 1983).Therefore we have proposed that creativity and Voice Quality are related. Creativity requires thinking, if an employee does not pay Attention to the organizational goals he/she cannot give an creative and good idea that is suitable for the organization (De Dreu, Baas & Nijstad, 2008). Since Attention is determined by the WMC of an employee, this enables the on-task Attention focus and prevents distraction of this task (de Dreu et al., 2012), we assume that differences in the WMC of employees will result in
creativity differences of the ideas voiced.
Furthermore, De Dreu, Baas and Nijstad (2008) developed the dual process theory towards creativity. They suggest that systematic scanning of cognitive categories using prolonged effort and time is a necessity to creatively explore a smaller number of categories. In other words; one way to be creative is to use a controlled systematic and persistence route which WMC fosters. We propose that not only creative thinking benefits from conscious Attention but also Voice Quality.
To sum up, it is stated that voice is a planned behavior and the outcomes are consciously calculated. Only this raises the question which role Attention and thus WMC plays in this behavior. If we assume that the WMC does play a role, the WMC of employees
who do voice will be different of the ones who do not. Yet, we have argued that the reason to voice or remain silent are emotion related and thus automatic decision making strategies. In line of reasoning, we predict that the WMC score of an employee (whether this is low or high) has no or little influence on the Voice Quantity of that employee and thus Attention does not play a role in this relationship. Which leads to the first hypothesis:
Hypothesis 1a: Attention has no influence on the amount of times a person voices,
particularly: Voice Quantity.
In contrast, the Voice Quality is argued to be influenced by creative abilities and conscious Attention focus. The ability to actively keep task relevant aspects in mind and to focus Attention is more likely to be of importance in the process of idea generation and thus Voice Quality. Differences in the WMC will thus play a role in the quality of the ideas that employees communicate. Which assumes that Voice Quality just as creativity is determined by systematic processing of employees. The second hypothesis therefore is:
Hypothesis 1b: The Attention of an employee positively influences the Voice Quality of
that employee.
1.4 Impulsivity and Voice Quantity vs Voice Quality
Attention is an example of a systematic cognitive system within individuals. Attention requires slow and conscious processing. While the automatic system is best described by processing that is stereotypical, speedy, global and unconscious (Evans, 2007), like for
example Impulsiveness. Individuals can be impulsive or non-impulsive. Impulsive individuals are described as: “Individuals who have the tendency to ‘lose their focus easily’ and are unable to control certain behavior” (Barratt, 1991). In other words (dys-)functional Impulsivity is related to speedy and non-reflexive decisions.
We distinguish these two types of Impulsivity because they have a different nature (Dickman, 1990). When people make a functional impulsive decisions this means that this strategy is appropriate for the situation and will have positive consequences. In contrast when making a dysfunctional impulsive decision the behavior that belongs to this decision is not appropriate and will have negative consequences (Vigil-Colet & Codomiu-Raga, 2004). The
consequences of Impulsivity are thus not always negative. Especially when a task is very simple or the time available to make a decision is really short, impulsive individuals are more accurate than non-impulsive individuals (Dickman, 1990).
As mentioned Impulsivity is a uncontrollable trait and can be impersonate itself in different situations (Barratt, 1991). Impulsivity thus is unconscious, it cannot be controlled by conscious Attention. Moreover, we have argued that Impulsivity is a personal trait that
influences individuals to make rapid decisions. In other words, impulsive individuals mostly make use of the automatic cognitive system (Dickman, 1990; Evans, 2007). Voice Quantity is a consequence of this system, the decision to voice is made in the automatic system (Detert & Burris, 2007; Edmondson, 2009). We therefore expect that highly impulsive individuals will mostly use the automatic system, which will influence the decisions the individual will make and can result in a decision to voice.
Furthermore, Impulsivity influences how individuals see their social context. For example: employees can feel the risk that speaking up will lead to a loss of support of their colleagues or supervisor and thus for instance restrict their career mobility (Detert & Burris, 2007). Risk-taking behaviors are those that involve a potential for danger or harm as well as a potential for some form of reward (Lejuez, Richards, Read, Kahler & Ramsey et al., 2002). To prevent potential negative outcome individual have the tendency to overlook the cost of the risk. This process requires Attention preceding the decision, which impulsive individuals lacks (Lejuez et al., 2002; Barratt, 1991). We therefore suggest that non-impulsive individuals
remain silent because of the social risk they are feeling to voice. While an impulsive
individual will only start to oversee the risk after making the decision to voice (Morrison & Milliken, 2003; Barratt, 1991). Impulsivity thus influences the decision making process in several ways. For this reason we suggest that an impulsive employee will Voice more often than a non-impulsive employee. This leads to the third hypothesis:
Hypothesis 2a: The impulsiveness of an employee will positively influence the Voice
Quantity.
Impulsivity is assumed to influence the automatic and thus unconscious system of an individual. Voice Quality on the other side is argued to be influenced by the systematic system (Wolsink, Forthcoming). Impulsive individuals do not use this system, as showed by the risk taking behavior of individuals (Morrison & Milliken, 2003; Barratt, 1991). We therefore suggest that an impulsive employee will voice more different ideas but this will make no difference in the quality in the things ideas voiced compared to a non-impulsive employee. Therefore having impulsive employees does not mean that the quality of ideas will go up, instead it can lead to a loss of information density because of the high amount of ideas (Morrison & Milliken, 2003). The fourth hypothesis is:
Hypothesis 2b: Voice Quality is not influenced by the impulsiveness of an employee.
1.5 The interaction effect of Attention and Impulsivity
To identify the perfect employee, the second goal of this paper, we suggest that there are characteristics that are more important for employees than others. In this study we assume that to voice high quality ideas, employees should be able to systematic process the available information (Wolsink, Forthcoming). In contrast, we assume that it is the employees automatic processing mode which will lead to an increase in the voice frequency (Barratt,
1991; Evans, 2007). So, separately the automatic and systematic systems are assumed to only influence the component of voice to which it is related.
The study of Dickman (1990) explains that Impulsivity does not necessary has
negative consequences. There are circumstances that Impulsivity can lead to positive problem solving. High impulsive individuals are more accurate than non-impulsive individuals, when a task is simple or the available time to make a decision is really short (Dickman, 1990).
Furthermore, we have argued that the current environment is fast and dynamic (Edmondson, 1999). For an employee the automatic system, which enables fast and automatic decision making, is thus important to keep up with the environment. Unfortunately, the automatic system will not lead to creativity and thus qualitative good ideas in itself.
Impulsivity’s consequences can be positive when the decision is based on past
experiences (Dickman, 1990). For example: in the past a decision needed conscious Attention and focus but over the years the employee has learned the consequences and developed it into automatic decision behavior. This suggests that an employee needed Attention and conscious processing for the same decision in the past. An employee should be able to learn for
Impulsivity to lead to positive results. Individuals thus develop their automatic decision making by learning behavior (Lejuez et al., 2002).
Furthermore, learning behavior requires Attention (Evans, 2007), and conscious Attention leads to creative idea generation (Wolsink, Forthcoming). For this reasons, we expect that an employee that is both high in Attention and Impulsivity will voice an higher amount of ideas which are also high in Quality than an employee who is just high in one of the two cognitive systems. Additionally, we suggest that this ratio will improve overtime trough learning behavior. The last two, hypotheses therefore are:
Hypothesis 3a: The Attention of an employee positively moderates the effect of
Impulsivity on Voice Quantity.
Hypothesis 3b: The Impulsivity of an employee positively moderates the effect of
Attention on Voice Quality.
2. Method
2.1 Design and Sample
The hypotheses of this study are designed to test whether Attention predicts the Voice Quality and Impulsivity predicts Voice Quantity of employees towards their manager and furthermore if there is an interaction effect between Impulsivity and Attention. For this reason data is collected through a survey and an additional WMC test (which measured Attention). The total sample contains 141 dyads of managers and employees from different companies throughout the Netherlands. Only Dutch speaking employees and supervisor were able to participate. Each thread consists of one manager and two employees, who work together on a regular basis. The manager only had to complete a survey. Both employees first had to do a WMC test after which they had to complete the survey. Using the purposive sampling technique, the data is collected by a research team of seven master students to acquire a larger reach of the population within the available time (Saunders et al., 2009, p. 365). There was no restriction in the kind of firms and sectors participating.
For this study the demographics age, gender, type of organization and duration of employment are important and thus should be a good representation of the population. The focus group is the labor force of the Netherlands. Therefore persons younger than 15 and older than 65 are left out of the study (CBS, Arbeid en sociale zekerheid; Beroepsbevolking, 2014). The average age of the managers is 42 years (SD=11.889) and the average age of the employees is 34 years, ranging from 16 to 65 years. Furthermore, out of the employees 51% is female and 49% male and out of the managers 36% is female and 64%. Lastly, only Dutch speaking workers were invited to participate in the questionnaire and experiments. Most participants work in the private sector (40%) and the duration of the employment was at average higher for the manager (135 months) than for the employees (83 months).
2.2 Procedure, test and measurements
Data collection was done for the whole dryad in one session. First contact was made with one of the three participants, in order to assure that the company and the other two participants were willing to participate and that it was allowed. Prior to testing, participants were informed with an information brochure about the procedure. Furthermore, the participants were
approached a day prior to the session. This to communicate their participant number and as a reminder to the session. At the day of the session participants were given a short introduction after which the employees were seated in a quiet room and started with the WMC test after which the survey followed. In the meantime the supervisors completed the survey in their own office.
The data was collected in a period of three months. At the employee level the independent variables Attention and Impulsivity were measured. The dependent variables: Voice Quantity and Voice Quality of the employee were measured at the manager level.
2.2.1 Dependent variables: Voice Quantity & Voice Quality
Voice Quantity and Voice Quality are measured through a questionnaire. The Voice Quantity and Voice Quality scales are not validated yet by former research. For this reason Wolsink (forthcoming) pretested the statements in order to assure validity and reliability. Supervisors should answer the questions according to their perception of the fit with the items and the Voice Quantity/Quality behavior of their employees, as this study focusses on the Voice Quantity/Quality of the employees perceived by their supervisor.
The Voice Quantity questionnaire is based on the work of Wolsink (Forthcoming). It includes three types of Voice Quantity: suggestion-focused voice, opinion-focused voice and problem-focused voice. The variable is measured at the supervisor level and the total
questionnaire includes 12 items, the three types of Voice Quantity equally divided. An
example item of suggestion focused voice is: ‘How often does your subordinate suggest solutions for problems within the company?’. An example item of opinion focused voice is: ‘How often does your subordinate gives his/her opinion about cases within the company?’. Lastly, an example item of problem focused voice is: ‘How often does your subordinate speaks out his or her worries about processes which do not function optimally’. Participants responded on a 1-5 Likert scale. A 1 meaning ‘never’ and a 5 meaning ‘often’.
The questionnaire for Voice Quality is based on the theory of Wolsink (Forthcoming) too. This questionnaire includes 24 items, of which sixteen are based on the originality and eight on usability. Participants responded on a 1-7 Likert scale. A 1 means ‘total
disagreement’ and a 7 means ‘total agreement’. An example question of the Originality questionnaire is: ‘If your subordinate comes with a suggestion, this is innovative’. Second, an example item of the utility questionnaire is: ‘The ideas of my subordinate are very useful’. The variable was measured at the supervisor level. In this questionnaire a difference is made between suggestion focused-, opinion focused- and problem focused voice too.
2.2.2 Independent variable: Attention
Attention (WMC) of the employee was measured for each employee separately. The researcher visited the employee at their work and administered the WMC test on a laptop, made available by the University of Amsterdam, in a quiet room. The participants first
received an oral explanation accompanied with a textual description for clarity. To ensure that the participant understood everything, the test started with eight trial exercises after which the actual test started when the participant was ready. The experimenter was present in the room during the whole session.
The Working Memory Capacity task was measured through eight trials. Participants were presented green rectangles or circles. When they were presented rectangles the purpose
was to recall the rectangle that was changed in the direction presented (horizontal to vertical; vertical to horizontal). When participants were presented circles the purpose was to recall the circles in a second empty screen presented right after the first presentation. Circles were presented simultaneously or one by one. Furthermore, there were yellow distractions included. For the rectangles these distractions were presented simultaneous or after the green rectangles and participants were intended to recall the changing green rectangle. Additionally, for the circles the distraction were presented: simultaneous with the green circles, all in once after the
green circles were presented (participants were presented three screens), or alternately serials
of one green circle, one yellow circle and so on (Sligte, Forthcoming).
Each of the trials repeated itself in a more complex manner as long as the participant had less than two errors in that trial. In total the participant was allowed to make sixteen mistakes (Sligte, Forthcoming). The WMC is a computerized, laboratory-based measurement, and describes the capacity of a person to keep task relevant information active in his/her mind. For this reason each stimulus remained on screen for 0.5 seconds after which the second screen was presented and during which the participant had to keep the information in his/her mind.
2.2.3 Independent variable: Impulsivity
Impulsivity is measured through a questionnaire. Former research has validated the scale for impulsiveness. Answers should have been given based on the employee’s perception of the fit of the item with their own behavior.
The questionnaire of Impulsivity is based on the theory of Barratt (1994). An example item is: ‘I take fast decisions’. This questionnaire includes 5 Impulsiveness items and 4 concentration items. Furthermore the questionnaire of impulsivity includes items based on the theory of Carver & White (1994) about behavioral activation scales. We think some of these
questions are in common with Impulsivity, this will be controlled by measuring the Cronbach alpha. An example item is: ‘I often do things just for fun’. Participants responded on a 1-7 Likert scale. A 1 means ‘total disagreement’ and a 7 ‘total agreement’. The total questionnaire included 5 impulsiveness, 4 concentration and 4 Bas fun seeking items and was measured at the employee level.
2.2.4 Control variables
We included two control variables: positive activating affect of the manager and age of the employee. First, age because the eager to learn and change the organization may decrease over time and thus the voice behavior of older employees too (Lejuez et al., 2002). Second, positive activating affect is a control variable because the research of Schwarz & Clore (1983) suggests that individuals may us their affective states on reports and judgments of general well-being. Moreover, Forgas (1995) suggests that there are affect states that have a priming effect, which means that the state of affect may indirectly influence the judgment of other situations as well. Baas, De Dreu and Nijstad (2008), explained that for example happiness is such a priming affect state and positively activates an individual’s judgment. We therefore suggest that managers that felt happy, had more positive judgments about their employees Voice Quantity and Voice Quality than mangers who were having more negative feelings. This may influence the results of this study.
2.3 Analysis and predictions
The data is imported in the computer program SPSS and the two moderation models are tested. Both moderation models comprise three models based on the predicted relationships between the variables. Therefore in total there are six models to answer the six hypotheses mentioned in the introduction.
In the first moderation model, Voice Quantity, model 1 tests the prediction that Attention has little or no effect on Voice Quantity. Model 2 tests the predicted positive effect of Impulsivity on Voice Quantity. Lastly, model 3 tests the predicted positive interaction effect between Attention and Impulsivity on the Voice Quantity.
In the second moderation model, Voice Quality, model 1 tests the prediction that Attention has a positive effect on Voice Quality. Model 2, tests the prediction that Impulsivity has little or no effect on Voice Quality. And last, model 3 tests the predicted positive effect of the interaction between Attention and Impulsivity on Voice Quality.
Lastly, both moderation models include the control variables age of the employees and positive activating affect of the manager.
3. Measurement models
3.1 Moderation model: Voice Quantity
3.2 Moderation model: Voice Quality
4. Results
4.1 Descriptive
141 triads out of 158 completed the questionnaire. Incomplete surveys with missing values on key variables were deleted. For other missing data, hotdeck was used. Before the regression analysis we ran a factor analyses on three of the four variables (Attention was excluded, this variable was tested by a task) to make sure that the variables only measured the factors that were important for this study.
4.1.1 Factor and reliability analysis
As expected, Voice Quantity to managers included three factors: Suggestion-focused, Problem-focused and Opinion focused. All composed of five items. Additionally the
Cronbach’s apha of this scale was 0.91. Second, Voice Quality to managers included the two expected factors: originality and usability. We started with 26 items, of these two were removed due to the fact that they did not fall under the measured factors. After this the
originality factor included sixteen items and the usability factor eight items. Cronbach’s alpha for this variable was 0.96. And last, the impulsivity scale included the three expected factors (Impulsivity, Concentration and BAS fun seeking). The Impulsivity factor included seven items and both the concentration and BAS factor included three items. Furthermore, the Cronbach’s alpha of Impulsivity was 0.82.
We chose not to remove the counterbalanced items due to the fact that after checking the Cronbach’s alpha, reliability was higher with these items in the scales. Furthermore, the Cronbach’s alphas were high for all variables and there was no opportunity to increase them by removing an item. The factor analyses resulted in four complete scales which are reflected in Appendix C. The Cronbach’s alpha for each variable is reflected in table 1.
4.1.2 Correlations & Control variables
The correlations and control variables of this study are stated in table 1. Most of our
expectations are confirmed by the correlations between the variables, but not all of them. As expected Attention (WMC) did positively correlate to Voice Quality ( r=0.194, P<.01), and did not correlate to Voice Quantity (r=0.138, ns). This means that employees with high Attention get higher ratings for the quality of voice but not for the quantity of voice by their managers. This indicates that indeed, systematic processing is more important for Voice Quality than for Voice Quantity. Furthermore, the expectation that Impulsivity would not correlate to Voice Quality is also confirmed (r=0.0.83, ns). However, Impulsivity did not correlate to Voice Quantity (r=-0.008, ns) which is unexpected. This means that employees who rate themselves as impulsive were not evaluated higher on quantity of voice by their managers than non-impulsive employees. Besides, age of the employee (Voice Quantity:
r=0.169, p<.05) and positive activating affect of the manager (Voice Quantity: r= 0.221,
p<.01; Voice Quality: r=0.346, p<.01) are control variables because of their relative high and
significant correlations with the main variables.
Table 1. Descriptive and correlations (Cronbach’s Alpha on the diagonal)
Note. N = 141. * p<0.05. ** p<0.01. 5 & 6 are control variables. Positive activating affect is
of the manager. M SD 1 2 3 4 5 6 1. Voice Quantity 3,120 0,688 (0,914) 2. Voice Quality 4,274 1,023 0,754** (0,955) 3. WMC 40,180 12,664 0,138 0,194* -4. Impulsivity 3,924 0,812 -0,008 0,083 0,117 (0,820)
5. Positive activating affect - - 0,221** 0,346** 0,009 -0,027 () 6. Age of the employee 33,660 12,246 0,169* 0,108 -0,327** -0,277** -0,134 ()
4.2 Regression analysis
4.2.1 Regressions and moderation model: Voice Quantity
Table two contains the results of three regression models for Voice Quantity. The first two models tested the two predicted direct effects for Impulsivity and Attention (WMC) on Voice Quantity. Employees’ WMC did not affect Voice Quantity as expected (β =0.137, ns, R² =0.019). Furthermore, the model only explained 2 % of variance. Impulsivity on the other hand did not have a positive effect on Voice Quantity (β = -0.009, ns, R² =0.000). Our
hypothesis that impulsive individuals voice more often is therefore rejected. This indicates, as expected, that respondents with a high WMC and thus Attention do not voice more than respondents low in Attention. Furthermore this indicates that this study did not find evidence for the prediction that impulsive respondents voice more often that non-impulsive
respondents.
The third model tested the predicted moderation effect of Attention in the relationship between Impulsivity and Voice Quantity using Process (Hayes, 2012). Process is an existing computational tool for estimating and probing various types of mediation, moderation, and conditional mediating modeling, provided by Preacher and Hayes (Cole, Bruch and Walter, 2008). In contrast with the expectations, the third model did not show a significant positive interaction effect of Impulsivity and Attention (β =0.005, ns). The explained variance of this model is 3 %. Probably due to the fact that there was no main effect of Impulsivity on Voice Quantity and no significant correlation between these variables was found as well. We found no evidence for the prediction that impulsive respondents who are high in Attention voice more than impulsive respondents who are low in Attention.
Table 2. Regression results Voice Quantity
Note. N=141. * p<0.05. ** p<0.01. *** p<0.001. Controlled for positive activating affect of the manager and age of the employee.
Model 1
Model 2
Model 3
Dependent Variabele
Voice Quantity
Voice Quantity
Voice Quantity
Coefficient
SE
Beta
Coefficient
SE
Beta
β
SE
T
Constant
2,818***
0,194
-
3,151***
0,291
-
3,711***
0,974
3,811
WMC
0,007
0,005
0,137
-0,012
0,023
-0,537
Impulsivity
-0,008
0,073
-0,009
-0,234
0,250
-0,936
WMC*Impulsivity
0,005
0,006
0,889
Positive activating affect
0,204**
0,064
0,256**
0,201**
0,066
0,253**
0,198**
0,066
0,248**
Age of the employee
0,016**
0,005
0,280**
0,013*
0,005
0,225*
0,012*
0,005
0,206*
F toets
2,639
0,012
1,1895*
R
²
0,019
0,000
0,025
4.2.2 Regression and moderation model: Voice Quality
The third table contains the results of the three regressions models for Voice Quality. The first two models tested the two predicted direct effects for Attention (WMC) (β =0.212, p <.01, R² =0.208) and Impulsivity (β =0.057, ns, R² =0.003) on Voice Quality. The results confirmed the expected main effects. Attention did affect Voice Quality and Impulsivity did not. The explained variance of the Attention effect was 21 % and is relatively high. This indicates that respondents with a high WMC score voice more useful ideas than respondents with a low WMC score. Furthermore, a respondent who rated himself as impulsive did not have a higher or lower Voice Quality than a respondent who was not-impulsive.
The third model, tested through process (Hayes, 2012), unexpectedly did not show a positive interaction effect of Attention and Impulsivity (β =-0.002, ns). This rejects the hypothesized moderation effect of Impulsivity on the relationship between Attention and Voice Quality. The explained variance of this models is 4 %. Probably due to the fact that there was no main effect of Impulsivity on Voice Quality and no significant correlations were found between these variables as well. This means that the prediction that respondents with a high Attention and who are impulsive voice more qualitative good ideas, than respondents high in Attention who are not Impulsive is not proved by the results of this study.
Table 3. Regression results Voice Quality
Note. N=141. * p<0.05. ** p<0.01. *** p<0.001. Controlled for positive activating affect of the manager and age of the employee.
Model 1
Model 2
Model 3
Dependent variable
Voice Quality
Voice Quality
Voice Quality
Coefficient
SE
Beta
Coefficient
SE
Beta
β
SE
T
Constant
3,576***
0,277
-
3,973***
0,420
-
4,068**
1,435
2,836
WMC
0,017**
0,007
0,212**
-0,002
0,034
-0,061
Impulsivity
0,070
0,105
0,057
-0,106
0,369
-0,289
WMC*Impulsivity
-0,002
0,034
0,519
Positive activating affect
0,430***
0,089
0,374***
0,428***
0,092
0,372***
0,419***
0,092
0,364***
Age of the employee
0,018**
0,007
0,225**
0,014*
0,007
0,169*
0,011
0,007
0,134
F toets
6,426**
0,441
2,0641*
R
²
0,208
0,003
0,043
4.3 Processing modes graphic against Voice Quantity & Voice Quality
Last, in figure 1 we have positioned Attention (facilitated by the systematic processing mode) and Impulsivity (facilitated by the automatic processing mode) against both Voice Quantity and Voice Quality. The results showed that Attention and thus the systematic processing mode only facilitates Voice Quality and not Voice Quantity. Unexpectedly, the results did not show that Impulsivity and thus the automatic processing mode facilitates Voice Quantity. However, the results did show that Impulsivity does not facilitate Voice Quality as expected. We reflected the Beta of the effects between the variables in figure 1.
Figure 1: The processing modes and Voice Quantity & Voice Quality
A. Voice Quantity B. Voice Quality
Note: A. Shows the beta of the direct and moderating effects on Voice Quantity. B. Shows the
beta of the direct and moderating effects on Voice Quality.
5. Discussion
The main goal of this study was to distinguish the two components (quantity and quality) of voice behavior, by linking them to the two different cognitive processing modes individuals have. We have tried to link Voice Quantity and Voice Quality to automatic and systematic processing systems. Moreover, we have argued that Voice Quantity relies more on
unconscious and automatic processing, while Voice Quality requires more conscious and systematic processing. To test this we linked Attention (Conscious) and Impulsivity
(Unconscious) to both Voice Quantity and Voice Quality. The research question of this study therefore was: What is the role of the two different processing modes (automatic &
systematic) and thus Attention in relation to the frequency of idea communication (Voice Quantity), versus the content of idea communication (Voice Quality)?
5.1 Link between theory, hypotheses and results
According to the research of Wolsink (Forthcoming) and Kane & Engle (2003), Attention is a systematic process which makes creative idea generation possible, and is furthermore central to WMC (Kane & Engle, 2003). The WMC of an employee is his/her capacity to selectively maintain and manipulate goal-relevant information without getting distracted (Jha et al., 2010). Furthermore, the literature assumes that Voice Quantity is influenced by the decision to voice. Decision making is influenced by emotions and risk taking tendencies and is
moreover a rapid and automatic process (Detert & Burris. 2007; Evans, 2007). We suggested that conscious Attention and keeping a large amount of information active in mind (WMC), is of limited influence in this part of the voice process. The first hypothesis, Attention has no
influence on the amount of times a person voices, particularly: Voice Quantity, was confirmed
by the results, which showed no relation between Attention and Voice Quantity.
Additionally, voicing is driven by the desire to change the status quo into something better. Only ideas that are original and useful have the potential to improve the status quo. So, in order to cause change, idea generation needs creativity (Amabille, 1983; De Dreu et al., 2010). Voice Quality is suggested to be based on idea generation (Wolsink, Forthcoming). Therefore, creativity and Voice Quality are suggested to be closely related. The literature, furthermore, suggests that creativity requires systematic scanning of cognitive categories using prolonged effort and time to creatively explore a smaller number of categories (De Dreu et al., 2008). Therefore this study has proposed that not only creative thinking benefits from conscious Attention (WMC) but Voice Quality (the communication of new and useful ideas) does too. We expected that Attention would positively influence Voice Quality because they are both linked to the systematic and conscious cognitive system. The second hypothesis: The
Attention of an employee positively influences the Voice Quality of that employee, was also
confirmed by the results. We found that employees with high Attention (who are able to systematically process information) voice higher quality ideas to their managers than employees with low Attention (hence, are less able to engage in systematic processing).
In contrast, the third and fourth hypothesis are based on studies linking the decision making process to the automatic cognitive system (Evans, 2007; Detert & Burris. 2007; Barratt, 1991). Impulsivity is a behavioral trait which is uncontrollable (unconscious) and speedy and can reveal itself in different situations. Individuals who have this trait are influenced to make rapid decisions (Barratt, 1991). The automatic system is described by personal characteristics that are speedy, stereotypical and unconscious (Evans, 2007).
Moreover, Impulsiveness influences an individual’s risk-taking behavior. To prevent potential harm, individuals have the tendency to overlook the cost of the risk, this process requires Attention which impulsive individuals lack (Lejuez et al., 2002; Barratt, 1991). For this reasons this study has proposed that Impulsivity is an unconscious behavior which is part of
the automatic processing mode of individuals. The overall suggestion of this study was that automatic systems influence each other. We therefore expected that Impulsivity would positively influence Voice Quantity because both the trait and the behavior are linked to the automatic and unconscious cognitive system. The third hypothesis was: The impulsiveness of
an employee will positively influence the Voice Quantity. And the fourth hypothesis: Voice
Quality is not influenced by the impulsiveness of an employee. Unfortunately we did not find
confirming results. However, this might be the result of methodological issues.
In this study impulsivity was measured by a questionnaire, which requires conscious Attention and moreover the consciously deliberation about their proposed unconscious actions of respondents. Since the theory suggest that Impulsivity is a unconscious behavior that individuals are not always aware of (Barratt, 1991), we suggest that perhaps, a questionnaire is not the best way to measure this. In other words, we have tried to measure something unconscious by asking for conscious evaluations of behavior. Additionally, some respondents may lack the insights or ability to provide an accurate self-report of their own behavior (Ladouceur, Bouchard, Rheaume, Jacques & Ferland, 2000). This knowledge strengthens the suggestion that it was the measurement that was wrong and not the theory. Future research should therefore try to improve this measurement of automatic systems.
A recommendation for this can be to replace the Impulsivity-scale by a risk taking test similar to Impulsivity, risk taking is described as an automatic system by Lejuez, Read,
Kahler, Richards, and Ramsey (2002). Risk taking can be measured with the BART (Balloon Analogue Risk Task). BART is a computerized, laboratory-based measurement that involves actual risk behavior for which, similar to real-world situations, risk taking is rewarded up until a point at which further risk taking results in poorer outcomes (Lejuez et al., 2002). The limitations of relying solely on self-report measurement can be overcome by a task like this. An experiment will identify the behavioral pattern of an individual without relying on the
self-report evaluations of this individual. Therefore a task is more likely to be a valid measurement of unconscious behavior (Lejuez et al., 2002).
The second more practical goal of this study was to identify the perfect employee. For this reason we tested the interaction effect of Impulsivity and Attention in relation to both Voice Quantity and Voice Quality. We expected that, because Impulsivity influences the decision to voice and Attention influences the quality of ideas voiced, a combination of these two characteristics would result in the perfect employee. This perfect employee was defined as an individual who would voice the most ideas that are high in quality. The last two hypotheses were: The impulsivity of an employee positively moderates the effect of Attention
on Voice Quantity; and The impulsivity of an employee positively moderates the effect of
Attention on Voice Quality. As well as the Impulsivity hypotheses these hypotheses were not
confirmed by the results of this study.
Besides the methodological issues of Impulsivity, another explanation for the not confirming moderation results is that it requires learning behavior (Lejuez et al., 2002). In the introduction we have argued that Impulsivity could have positive consequences when the decision is based on past experience (Dickman, 1990). Dickman (1990) suggested that when a decision in the past needed Attention and an employee had used this Attention to make the decision, the employee has learned the consequences of the decision overtime and can develop automatic decision making behavior for similar problems. Learning can thus reduce the importance of Attention in future decision making and can result in positive outcomes of Impulsivity. Additionally, we have argued that learning takes time and thus the suggested interaction effect will improve overtime (Lejuez et al., 2002). An repetition of this study over a set timeframe therefore could lead to different results that might confirm the suggested interaction effects.
5.2 Influences of the design of the study and other discussion points
The data collection of this study had to be done on location. Availability of internet and a separate room was a necessity too. This excluded several organizations because of distance or unavailability of the necessities. Additionally, the time frame of the data collection was relatively long (on average two hours), this made it less attractive for organizations to cooperate. On the other hand, the researchers could check if everything went correctly, this resulted in a high percentage of complete questionnaires. To improve this study and get more accurate data the duration of the data collection should be shortened. Which can be realized for example by introducing the suggested BART task instead of part of the questionnaire.
Furthermore, this study has no possibilities to generalize the findings across cultures. Only Dutch speaking managers and employees were invited to take part and respondents thus were western in general. Additionally, we did not state if employee-manager relationships were close. Closer relationships could have result into influenced answers of the manager and backwards.
Moreover, there are two bias problems with questionnaires. First, people are biased when they evaluate themselves, but others as well (Podsakoff & Organ, 1986). This can occur because the information given in the questionnaire is perceived different by the respondents but also due to social pressure. Social pressure can lead to dishonest answers because respondents give answers they think they should instead of their true opinions. Ideally one employee should be measured by more managers than just one in order to reduce this bias. Second the sample can be biased (Podsakoff & Organ, 1986). The researchers recruited from their social network and furthermore distance limited the sampling environment. The sample thus might not be truly random like supposed.
However, this study included different sectors and industries, non-profit and profit organizations. This resulted in a relative large variety in the respondents of the study.
Moreover, commitment of the respondents was established because they actually have meet the researchers. This led to a high amount of complete questionnaires because of the
established goodwill. Furthermore, we made use of control variables (Age of the employee & positive activating affect of the manager) which other studies have not. This made our study more complete. For example it is proposed that a happy manager will rate his employees more positive than a depressed (negative activating affect) manager. This will influence the
evaluation of Voice Quantity and Voice Quality and thus the results. Lastly, Attention was measured by a WMC task which is an objective measurement (Lejuez et al., 2002). The subjectivity issues accomplished with the Impulsivity-scale were therefore dismissed for this measurement and results are assumed to be highly objective.
5.3 Conclusion & Contribution
The research question, what is the role of the two different processing modes (automatic & systematic) and thus Attention in relation to the frequency of idea communication (Voice Quantity), versus the content of idea communication (Voice Quality), could not be answered completely. The suggestion that the components of voice behavior require two different processing modes is strengthened by the positive effect of Attention on Voice Quality and the fact that Attention did not predict Voice Quantity. However, for a more in depth insight into the automatic processing effects, further research in needed.
The theoretical contribution of this research is that we further differentiated the components of voice (Quantity & Quality), by linking Attention (systematic processing) to Voice Quality. The systematic processing mode only influenced Voice Quality, as expected, and did nothing for Voice Quantity. The suggestion that Voice Quality requires conscious Attention and thus is part of the systematic and slow cognitive system is therefore
strengthened.
The practical contribution for managers is that they should focus on the Attention of their employees because this is of most importance if managers want their employees to voice qualitative ideas. Attention partly is in the nature of employees, but when employees engage repeatedly in for example mindfulness training this process can be strengthened (Jha et al., 2010). Mindfulness is a mental mode characterized by full Attention to present-moment experience and mindfulness training programs offer exercises and guidance to help employees cultivate this Attention (Jha et al., 2010). We therefore suggest that if managers want to fully use the Attention of their employees that they facilitate such training programs for their employees. Even more, because we suggest that Attention also is a necessity for learning behavior, which is argued to be important for Impulsivity to result in positive outcomes, in this case a high frequency of voicing which is also high in quality.
References
Amabile, T. M. (1983). The social psychology of creativity: A componential
conceptualization. Journal of Personality and Social Psychology, 45, 357-376. Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of
judgments. Groups, Leadership, and Men. S, 222-236.
Ashford, S. J., & Black, J. S. (1996). Proactivity during organizational entry: The role of desire for control. Journal of Applied Psychology, 81(2), 199-214.
Baas, M., De Dreu, C. K. W., & Nijstad, B. A. (2008). A meta-analysis of 25 years of mood-creativity research: hedonic tone, activation, or regulatory focus? Psychological
Bulletin, 134(6), 779-806.
Baddeley, A. (2003). Working memory: Looking back and looking forward. Nature
Neuroscience, 4, 829-839.
Barratt, E. S. (1991). Measuring and predicting aggression within the context of a personality theory. Journal of Neuropsychiatry, 3, 535–539.
Barratt, E. S. (1994). Impulsiveness and aggression.Violence and mental disorder
Developments in risk assessment,10, 61-79.
Barrett, L. F., Tugade, M. M., & Engle, R. W. (2004). Individual differences in Working Memory Capacity and dual-process theories of the mind. Psychological Bulletin,
130(4), 553-573.
Bateman, S., & Crant, J. M. (1993). The proactive component of organizational behavior: A measure and correlates. Journal of organizational behavior, 14(2), 103-118.
Bindl, U. K., & Parker, S. K. (2010). Proactive work behavior: Forward-thinking and change-oriented action in organizations. APA handbook of industrial and organizational
psychology 2, 567-598.
CBS (2013). Arbeid en sociale zekerheid; Beroepsbevolking. Access: 31-01-2014. (http://www.cbs.nl/nl-NL/menu/themas/arbeid-sociale
zekerheid/publicaties/barometer-beroepsbevolking/barometer-werkzame beroepsbevolking-art.htm).
Cole, M.S., Walter, F., & Bruch, H. (2008). Affective mechanisms linking dysfunctional behavior to performance in work teams: A moderated mediation study. Journal of
Applied Psychology, 93(5), 945-958.
Conway, A. R. A., Cowan, N., & Bunting, M. F. (2001). The cocktail party phenomenon revisited: The importance of working memory capacity. Psychonomic Bulletin &
Review, 8(2), 331-335.
Crant, M. J. (2003). Speaking up when encouraged: Predicting voice behavior in a naturally-occurring setting. Paper presented at the annual meeting of the Academy of
Management, Seattle.
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.
De Dreu, C. K., Baas, M., & Nijstad, B. A. (2008). Hedonic tone and activation level in the mood-creativity link: toward a dual pathway to creativity model. Journal of
personality and social psychology, 94(5), 739.
De Dreu, C. K. W., 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 and Social Psychology Bulletin, 38(5), 656-669.
Dickman, S. J. (1990). Functional and dysfunctional impulsivity: Personality and cognitive correlates. Journal of Personality and Social Psychology, 58(1), 95-102.
Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams.
Administrative Science Quarterly, 44, 350−383.
Edmondson, A. C. (2003). Speaking up in the operating room: How team leaders promote learning in inter disciplinary action teams. Journal of Management Studies, 40, 1419-1452.
Forgas, J. P. (1995). Mood and judgment: the affect infusion model (AIM). Psychological
Bulletin, 117(1), 39-66.
Grant, A. M., & Ashford, S. J. (2008). The dynamics of proactivity at work. Research in
organizational behavior, 28, 3-34.
Green, L., Myerson, J., & McFadden, E. (1997). Rate of temporal discounting decreases with amount of reward. Memory & Cognition, 25, 715–723.
Harrison, T. L., Shipstead, Z., Hicks, K. L., Hambrick, D. Z., Redick, T. S., & Engle, R. W. (2013). Working memory training may increase working memory capacity but not fluid intelligence. Psychological science, 1-11.
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from (http://www.afhayes.com/public/process2012.pdf).
Hinson, J. M., Jameson, T. L., & Whitney, P. (2003). Impulsive decision making and working memory. Journal of Experimental Psychology, 29(2), 298-306.
Hirschman, A. O. 1970. Exit, voice, and loyalty: Responses to decline in firms, organizations,
and states. Cambridge, MA: Harvard University Press
Jha, A. P., Stanley, E. A., Kiyonaga, A., Wong, L., & Gelfand, L. (2010). Examining the protective effect of mindfulness training on working memory capacity and affective experience. American Psychological Association, 10(1), 54-64.
Kane, M. J., & Engle, R. W. (2003). Working-memory capacity and the control of attention: The contributions of goal neglect, response competition, and task set to stroop interferenence.
Journal of Experimental Psychology: General, 132, 47-70.
Ladouceur, R., Bouchard, C., Rheaume, N., Jacques, C., Ferland, F., Leblond, J., & Walker, M. (2000). Is the SOGS an accurate measure of pathological gambling among children, adolescents and adults? Journal of Gambling Studies, 16, 1–24. Lejuez, C. W., Read, J. P., Kahler, C.W., Richards, J. B., & Ramsey, S. E. et al. (2002).
Evaluation of a behavioral measure of risk taking: The balloon analogue risk task(BART).
Journal of Experimental Psychology, 8(2), 75-84.
LePine, J. A., & Van Dyne, L. (2001). Voice and cooperative behavior as contrasting forms of contextual performance: Evidence of differential relationships with Big Five
personality characteristics and cognitive ability. Journal of Applied Psychology, 86, 326 -336.
Liu, W., Zhu, R., & Yang, Y. (2010). I warn you because I like you: Voice behavior,
employee identifications, and transformational leadership. The Leadership Quarterly,
21, 189-202
Milliken, F. J., Morrison, E. W., & Hewlin, P. F. (2003). An exploratory study of employee silence: Issues that employees don't communicate upward and why. Journal of
Management Studies, 40, 1453-1476.
Morrison, E. W., & Milliken, F. J. (2000). Organizational silence: A barrier to change and development in a pluralistic world. Academy of Management Re view, 25, 706-725. Morrison E. W., Milliken F. J. (2003). Guest editors’ introduction: Speaking up, remaining
silent: The dynamics of voice and silence in organizations. Journal of Management
Studies, 40, 1353–1358.
Morrison, E. W. (2011). Employee voice behavior: Integration and directions for future research. The Academy of Management Annals, 5(1), 373-412.
