The relationship between affect and voice behaviour
And the influences of different responses to
capitalization
Amsterdam, June 22, 2014
Thesis seminar Business studies Supervisor: R. van Geffen Academic year: 2013-2014 Semester 2, Block 3 Midas Snoek, 10181520
Abstract
Being well informed in business nowadays is important. Voice behaviour among employees can help companies to be well informed. Although moods of employees can play a crucial part in their
behaviours, no study has ever looked at the direct relationship between moods and voice behaviour. This study will also look at the influences of responses to capitalization attempts. We propose that affect influences voice behaviour and that responses to capitalization will change this relationship. More specifically, we hypothesize that affect has a direct influence on voice behaviour and that active-constructive responses positively enhance this relation and passive responses negatively enhance this relation. We tested these hypotheses with samples of 65 employees. Only the direct influence of positive affect on employee voice behaviour was found, suggesting that when employees are in a positive mood they are more likely to produce voice behaviour.
Content
Abstract ... 2
Foreword ... 4
1. Introduction ... 5
2. Literature review ... 6
2.1 Employee voice behaviour ... 6
2.2 Employee affect ... 9
2.3 Capitalization ... 11
3. Theoretical framework ... 13
3.1 Employee voice behaviour ... 13
3.2 Employee affect ... 13 3.3 Capitalization ... 15 4. Method ... 17 4.1 Design ... 17 4.2 Sample ... 18 4.3 Measurements ... 19
4.3.1 Dependent variable: employee voice behaviour ... 19
4.3.2 Independent variable: employee affect ... 19
4.3.3 Moderator variable: response to capitalization ... 20
4.4 Procedure ... 20
4.5 Analysis ... 21
4.6 Predictions ... 21
5. Results ... 22
5.1 Reliabilities and correlations ... 22
5.2 Results ... 23
6. Discussion ... 27
6.1 Summary of results ... 27
6.2 Unexpected results and alternative explanations ... 29
6.3 Discussion points ... 30
6.4 Contributions, practical implications and future research ... 31
6.5 Conclusion ... 33
References ... 34
Foreword
I wrote this thesis for my Bachelor degree in Economie en Bedrijfskunde at the Universiteit van Amsterdam. While writing this thesis I got help from certain people which I would like to thank. First of all I would like to thank my supervisor, Renske van Geffen, who helped me to make sure that my thesis was feasible but also a lot more challenging and complete. I would also like to thank Renske and my brother Lukas for the helping me with my statistical analysis, which I could not have done without them. Next, I would like to thank my mother who was a great help in spreading my survey among a different network of people. And finally I would like to thank all people who were willing to fill in the surveys needed to collect my data. Without you, my thesis would be a lot less interesting to read.
I hope you enjoy reading my bachelor thesis.
Midas Snoek
1. Introduction
Being omniscient in the business world is impossible in the fast changing and complex world of today. But being well informed is crucial for business to survive. So it is essential to have information from all levels in the company. Employees play a crucial part in spotting opportunities or threats in companies nowadays. That is why companies often want employees to talk constructively about ideas, suggestions or opinions with the aim of changing the organization, even if these ideas challenge the status or opinions of powerful others. This coming up with constructive ideas and suggestion is called voice behaviour (Van Dyne & LePine, 1998). Employee voice behaviour is thus important, and therefore extensively studied before.
Voice behaviour is important because it has enormous implications for an organization’s performance since it can highlight problems, it comes up with suggestions to solve these problems and is a way to improve the organization through ideas. Input from all levels of the organization is essential to respond to the dynamic business environment in which we live today (Morrison, 2011). The business environment today explicitly asks companies to be flexible, innovative and to create continuous improvements.
The existing knowledge about voice behaviour is gigantic. Lots of factors influence voice behaviour directly or indirectly. Factors like which motives stimulate people to produce voice behaviour or to remain silent (Grant & Ashford, 2008; Milliken, Morrison & Hewlin, 2003), the decision process whereby the individual considers the positive and negative consequences (Ashford, Rothbard, Piderit & Dutton, 1998; Detert & Burris, 2007; Morrison & Milliken, 2000) and the
influence of the type of leadership on employee voice behaviour (Liu, Zhu & Yang, 2010) are already studied. As stated before, voice behaviour is extensively studied in relationship with different factors. However, the amount of studies done about voice behaviour and different factors like affect and capitalization are scarce. That makes this study relevant in two ways. It is relevant in a scientific way that it will contribute to the literature about voice behaviour. More specifically in a relationship with affect and capitalization. The practical relevance is obvious. As the introduction suggests, it is vital for
organizations to be well informed. Voice behaviour can help to inform the company. In practice the information acquired by this study can help companies with producing more voice by stimulating employee moods and certain behaviours among co-workers.
This study will close the gap in the literature on the relationship between employee affect and employee voice behaviour in a context with capitalization. This results in the following research question: Do active-constructive and passive responses to employee capitalization influence the relationship between employee affect and employee voice behaviour?
2. Literature review
2.1 Employee voice behaviour
Voice behaviour is intensively researched through the years resulting in several definitions. The definition of Van Dyne and LePine (1998): Promotive behaviour that emphasizes expression of constructive challenge intended to improve rather than merely criticize. Making innovative
suggestions for change and recommending modifications to standard procedures even when others disagree. The definition of Detert and Burris (2007) in a more recent study: The discretionary provision of information intended to improve organizational functioning to someone inside the organization with the perceived authority to act, even though such information may challenge and upset the status quo of the organization and its power holders. Those different definitions have several important commonalities described in a review done by Morrison (2011). The first
commonality is that voice behaviour is an act of verbal expression between a sender and a recipient. The second commonality is that voice behaviour is discretionary behaviour which means that individuals can choose for themselves to produce voice behaviour. The third commonality is that voice behaviour is of a constructive nature which means that it is useful (Morisson, 2011). Morisson (2011) tried to create an integrated conceptualization of voice behaviour: Discretionary
communication of ideas, suggestions, concerns or opinions about work-related issues with the intent 6
to improve organizational or unit functioning. Now we know what the definitions of employee voice behaviour are we can take a closer look at what the phenomenon all contain.
Certain behaviours have been described in the past which are very closely related, or have certain characteristics, that we call voice behaviour today. Katz (1964) described that certain behaviours are necessary in a functioning organization. These behaviours were later called
organizational spontaneity by George and Brief (1992) and refer to behaviours not specified by role prescriptions, but that facilitate the accomplishment of organizational goals. Those behaviours were also called extra-role behaviour. Organizational spontaneity has five forms, namely: helping co-workers, protecting the organization, making constructive suggestions, developing oneself and spreading goodwill (Katz, 1964; George & Brief, 1992). Later Katz and Kahn (1978) stated that it almost never happens that the task of making creative suggestions for improving the functioning of the organization is assigned to employees. This makes one of the five forms of organizational spontaneity, making constructive suggestions, very closely related to voice behaviour. Hence, voice behaviour is also described as discretionary and with the main goal to improve (Van Dyne & LePine, 1998; Detert & Burris, 2007).
Organizational citizenship behaviour is also a phenomenon which has several characteristics that voice behaviour also possesses. Organ (1988) defined organizational citizenship behaviour as individual behaviour that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization. Organ (1988) described five forms of organizational citizenship behaviour, namely: altruism,
conscientiousness, sportsmanship, courtesy and civic virtue. Altruism and civic virtue contain characteristics of voice behaviour (George & Brief, 1992). Altruism includes discretionary behaviours that have the effect of helping a specific other person with an organizationally relevant task or problem (Organ, 1988). Civic virtue was first defined by Graham (1986) and later used by Organ (1988). It includes behaviours like attending meetings, discussing issues on personal time and speaking up. Altruism contains voice behavioural characteristics because it is voluntarily behaviour
which focusses on helping fix organizational problems. Civic virtue is related to voice behaviour in the fact that the characteristics of civic virtue contain speaking up. Speaking up is voice behaviour, when it is discretionary of nature. The related behaviours and voice behaviour itself all have some sort of content.
The content of voice is broad and studied thoroughly. According to Morrison (2011) there are four different contents of voice behaviour. It can be a way to improve a certain situation in a
company according to Van Dyne & LePine (1998). Miliken et al., (2003) state that it can be about an organizational or work-related problem. Pinder & Harlos (2001) describe that it can address a situation of misconduct or unfairness. And finally Premeaux and Bedeian (2003) suggest that it can be an opinion that is different from others. What people say when they voice and whether people voice at all is of course influenced by all sorts of factors.
A lot of research has been done on factors that influence voice behaviour. Those factors can be divided into four different categories according to Morrison (2011). Those categories are:
contextual factors, supervisor behaviour, employee attitudes and dispositions and other individual-level factors. Contextual factors can be summarized as factors that can have an important effect on the way in which employees are able to produce voice behaviour. The easier it is to communicate suggestions and ideas, the more likely people are to produce voice behaviour. Factors like the level of bureaucratic work to produce voice behaviour and the hierarchic levels in companies are important factors which are able to stimulate or inhibit voice behaviour (Glauser, 1984; Athanassiades, 1973). Supervisor behaviour focusses on the role of leader behaviour. A study from Ashford, Sutcliff and Christianson (2009) found that the more open the relationship between employees and their supervisor is, the more likely employees are to produce voice behaviour. But despite many studies there is still no really clear picture of how leaders influence employees to speak up or not (Morrison, 2011). Employee attitudes and dispositions are about differences between employees which has influence on the quantity they voice. An example is that employees who are satisfied at work are more likely to produce voice behaviour. In contrast employees who seem to be psychological
detached from their work are less likely to produce voice behaviour (Detert & Burris, 2007). Other individual-level factors are factors beyond attitudes and personality. Those are factors like tenure, which is positively related to voice behaviour (Tangirala & Ramanujam, 2008).
Concluding, a lot of research has been focussing on voice behaviour. Especially the factors that influence voice behaviour have been studied thoroughly. All four categories shortly described above give a good view of how voice behaviour gets influenced by different factors. One of the factors that has not been studied before is employee affect. Affect plays a central role in everybody’s life and therefore also in employees' life. This means that affect could be a big influence on whether employees speak up or not. It can be relevant to investigate this subject deeper. This will be done in this study.
2.2 Employee affect
Affect is a fairly broad term referring to a subjective feeling state that incorporates long-lasting mood states, such as cheerfulness or depression, as well as more specific ones, such as happiness or anger (Frijda, 1993). Frijda (1993) stated that moods and emotions are subtypes of affect. Emotions are, according to Frijda (1993), strongly directed toward a specific stimulus, for example a traffic jam or an unfriendly cashier. Moods on the contrary, are not directed toward a specific object (Parrott, 2001). Moods are a more general feeling which tend to be relatively enduring and pervasive and in general of lower intensity (Frijda, 1993).
A lot of studies in the past have focussed on self-reported mood but Watson and Tellegen (1985) stated that there was a lack of consensus concerning the dimensional structure of affect and did a reanalysis of a number of studies to come to more consensus. Two main dimensions have emerged from their studies: positive affect and negative affect (Watson & Tellegen, 1985). One is likely to think that those two dimensions are opposites but they are in fact independent uncorrelated dimensions according to factor analytic studies done by Watson & Tellegen (1985). Watson, Clark and Tellegen (1998) stated that the scales used to measure moods were never adequate and showed
low reliability. They created two scales which both contain ten measures which indicate a certain affect. For positive affect those are: enthusiastic, interested, determined, excited, inspired, alert, active, strong, proud and attentive. For negative affect those are: scared, afraid, upset, distressed, jittery, nervous, ashamed, guilty, irritable and hostile (Watson et al., 1998). Positive affect indicates the extent to which an individual feels active, alert and enthusiastic. A high score on positive affect means a high state of energy, full concentration and pleasurable engagement. A low score on positive affect means a state characterized by sadness and exhaustion (Watson et al., 1998). Negative affect indicates subjective distress and unpleasurable engagement. A high score on negative affect means an individual can experience anger, disgust and fear. A low score on negative affect can for example result in a state of calmness and serenity (Watson et al., 1998). Other studies investigated moods further.
Baas, De Dreu and Nijstad (2008) did a meta-analysis of mood-creativity research and made another division of affect. They divided different mood states to hedonic tone, activation and
regulatory focus. Hedonic tone refers to the valence of moods in which they can be either positive or negative (Baas et al., 2008). Activation refers to the extent in which people act actively. Positive activating moods are moods like happy and elated whereas positive deactivating moods refer to moods like relaxed (Baas et al., 2008). Regulatory focus refers to whether certain moods are
promotion focussed or prevention focussed. Promotion focus is originated from the survival need of nurturance and prevention focus is originated from the survival need of security (Higgins, 1997). According to the distribution to these three factors Baas et al. (2008) suggest that positive,
deactivating, prevention focussed moods are calm, serene and relaxed. This varies to what Watson et al. (1998), calmness and serenity are results of a low score on negative affect, stated. Moods are thus complex phenomenon which also have their influences.
Influences of affect are numerous and are studied thoroughly. As stated before, Baas et al., (2008) did a meta-analysis of the relationship between mood and creativity. They included 66 reports to come to more consensus in the relationship between mood and creativity. They looked at what
kinds of moods influenced creativity and found that activating moods stimulate creativity and that deactivating moods inhibit creativity (Baas et al., 2008). Another influence of moods is investigated by George in 1991 and 1992.
George (1991) argued that affective states are a direct indicator of behaviours and that is of interest for researchers to look at affective states and their influence on behaviour. George (1992) later studied the relation between positive moods and pro-social behaviour. He found that positive mood states can lead to more pro-social behaviour at work. This was in line with studies of Isen (1970, 1987) that suggested that people who experience positive affect often affiliate altruistic behaviours and exhibit helping tendencies. Also Baron studied the influences of moods in 2008. A more recent and more practice aimed study focussed on the influences of affect on the entrepreneurial process. This study, done by Baron in 2008, suggests that affect influences several aspects of the entrepreneurs’ cognition and therefore important elements of the entrepreneurial process. In his study, Baron (2008) concluded that affect influences many aspects of cognition and behaviour and that those influences are especially likely to occur at entrepreneurs because of their unpredictable and uncertain environment. It also influences many tasks entrepreneurs perform when launching new ventures. Through its impact on cognition, affect may have important influences on key aspects of the entrepreneurial process like opportunity recognition, the capacity to respond effectively in highly dynamic environments and success in acquiring needed resources (Baron, 2008). Concluding, affect has had a lot of attention from researchers in the past. Influences of affect are so numerous that including all in the literature review was impractical work. Although, while writing this literature review, one thing gets clear. What is not studied, is the relationship between affect and voice behaviour, therefore this relationship will be investigated in this study.
2.3 Capitalization
Capitalization is a term used to refer to when people seek out others to inform about a personal positive event and thereby deriving additional benefit from it (Gable, Impett, Reis & Asher, 2004).
Langston (1994) studied this field earlier due to lack of research done about when good things happen. Studies until then focussed on negative events and Langston (1994) suggested that it was because positive events in life are expected and because they do not call attention to themselves. Negative events often ask for active processes to ease their effects, whereas positive events are less likely to demand active responses (Taylor, 1991).
Lewinsohn and Libet (1972) and Lewinsohn and Graf (1973) studied positive events and their effects for the first time. They found that the occurrence of positive events was associated with less depressive affect within the day of the events’ occurrence. The conclusions of those studies were quite obvious: one is less depressed (less sad) on days when one is less depressed (laughs more) (Lewinsohn & Libet, 1972; Lewinsohn & Graf, 1973).
Just as Lewinsohn, Libet and Graf studied the effects of positive life events, so did Block and Zautra in 1981. They found a pattern suggesting that negative events tend to be related to increases in negative psychological outcomes but not to decreases in positive psychological outcome
measures. Positive events were related to increases in positive psychological outcomes but not to decreases in negative psychological outcomes (Block & Zautra, 1981).
As stated earlier, Langston (1994) studied capitalizing and coping with daily life events. In his first study he found that when people made expressive responses or felt like they were in control after a positive event, they felt greater positive affect and involvement than when they just had continued their routine. The already-positive effects of the positive events were enlarged when people responded expressively (Langston, 1994). In his second study he finds extra supporting evidence that above and beyond the quantity and valence of a day’s events, the expressiveness of a person’s reaction to them was a powerful predictor of their affective consequences.
Gable et al., (2004) studied the intrapersonal and interpersonal benefits of sharing of positive events. In line with Langston (1994) they found that on days that people communicated positive events to others, they experienced greater positive affect and satisfaction with life, above and beyond the importance of the event itself (Gable et al., 2004). Gable et al. also looked at different
ways to respond to people who shared positive events. They found that when people react in an active and constructive way and not in a passive or destructive way to the sharing of good news, it is associated with increased personal benefits above and beyond the effects of sharing the news itself. Concluding, capitalization (or the sharing of positive life events) has its influences on
important life factors such as affect. Those findings have contributed to the scientific knowledge about positive life events and the effect on people. This study will look at how this sharing of positive life events can influence the relation between affect and work behaviours. More specific voice behaviour.
3. Theoretical framework
3.1 Employee voice behaviour
Voice behaviour as defined by Van Dyne and LePine in 1998 is promotive behaviour that emphasizes expression of constructive challenge intended to improve rather than merely criticize. Making innovative suggestions for change and recommending modifications to standard procedures even when others disagree. The suggestions done by voicing of employees have the goal to improve organizational functioning (Morrison, 2011). The content of voice is fairly broad and can be put in three categories according to Morrison (2011). It can be about improving an organizational or work-related problem (Milliken et al., 2003), about a situation containing unfairness or misconduct (Pinder & Harlos, 2001) or about an opinion that differs from views of others (Premeaux & Bedeian, 2003). In this study there will be no differentiation made to the content of voice.
3.2 Employee affect
Frijda (1993) defined affect as the most general term, referring to a subjective feeling state that incorporates long-lasting mood states, such as cheerfulness or depression, as well as more specific ones, such as happiness or anger. Watson and Tellegen (1985) and Baas et al., (2008) distinguished
two hedonic tones in affect, positive and negative. The positive hedonic tone covers moods like calm, serene, relaxed, happy, upbeat and elated. The negative hedonic tone covers moods like sad,
discouraged, disappointed, anger and frustrated (Baas et al., 2007). These different mood states are related to organizational spontaneity according to George and Brief (1992). Organizational
spontaneity is an overlapping term covering organizational citizenship behaviour and pro-social organizational behaviour (George & Brief, 1992). Organizational citizenship behaviour is employee behaviour that goes beyond prescribed role requirements in its benefits to the organization
according to Katz & Kahn (1978). Later research stated that positive moods influence the quantity of organizational citizenship behaviour expressed by employees (George & Brief, 1992). However, as stated before, extensive research has been done about the relationship between affect and organizational spontaneity, the specific relation between affect and voice behaviour is not researched before. But because those two are related we state that employee positive affect stimulate employee voice behaviour, as can be seen in figure 1.
Hypothesis 1 [H1]: Employee positive affect stimulates employee voice behaviour.
Although the relationship between positive affect and organizational citizenship behaviour is thoroughly studied (George & Brief, 1992; Organ & Ryan, 1995; Williams & Shiaw, 1999; Bateman & Organ, 1983), the emphasis was never on negative affect and organizational citizenship behaviour. Only George and Brief (1992) suggested that negative moods could have a positive effect on the degree of critical thinking employees could express, which could eventually be an indicator of voice behaviour. Next to that, no research has been done about the relationship between negative affect and voice behaviour. We suggest that negative affect inhibits employee voice behaviour. Hypotheses can be seen in figure 1.
Hypothesis 2 [H2]: Employee negative affect inhibits employee voice behaviour
3.3 Capitalization
Capitalization is the process of informing another person about the occurrence of a personal positive event and thereby deriving additional benefit from it (Gable et al., 2004). Langston (1994) found that when people shared positive news with others or celebrated this positive news, they experienced greater positive affect beyond increases associated with the valence of the positive event itself. We conclude that capitalization efforts influence the affect of people sharing the positive event. Gable et al., (2004) researched different responses to these capitalization efforts. They researched four types of responses adapted from Rusbult, Zembrodt and Gunn (1982): constructive,
active-destructive, passive-constructive and passive-destructive. Responses are active when someone actually responds and passive when someone for example shrugs and walks away (Rusbult et al., 1982). Constructive responses contain positive affirming information and destructive responses lack this affirming information (Gable et. al., 2004).
As Langston (1994) and Gable et al. (2004) already found is that affect, satisfaction,
commitment, intimacy and trust were influenced when people did capitalization efforts. Gable et al. (2004) found that the different responses to these efforts had significant impacts on several factors like affect, commitment, satisfaction, intimacy and trust. They found that active-constructive
responses were positively correlated with those factors. Passive-constructive, active-destructive and passive-destructive were negatively correlated with those factors.
Because there is a positive correlation between active-constructive responses and affect and the fact that we hypothesised that positive affect leads to more voice behaviour, we come to the following hypothesis.
Hypothesis 3.1 [H3.1]: Active-constructive responses to employee capitalization enhance the positive relationship between employee positive affect and employee voice behaviour
In hypothesis 2 we suggested the negative relation between employee negative affect and employee voice behaviour. Because active-constructive responses to capitalization efforts are positively
correlated with affect and the fact that we hypothesised that negative affect leads to less voice behaviour we come to the following hypothesis.
Hypothesis 3.2 [H3.2]: Active-constructive responses to employee capitalization will positively influence the negative relationship between employee negative affect and employee voice behaviour
Both kinds of passive responses, as presented by Rusbult et. al. (1982), are negatively correlated to affect (Gable et. al., 2004). We suggest in hypothesis 1 that positive affect leads to more voice behaviour. So we hypothesise that passive responses to capitalization efforts will diminish the positive relationship between employee positive affect and employee voice behaviour.
Hypothesis 4.1 [H4.1]: Passive responses to employee capitalization will diminish the positive relationship between employee positive affect and employee voice behaviour
Based on hypothesis 2 and the construct of the negative correlation between passive responses and affect, we hypothesis that passive responses to employee capitalization will enhance the negative relationship between employee negative affect and employee voice behaviour.
Hypothesis 4.2 [4.2]: Passive responses to employee capitalization will enhance the negative relationship between employee deactivating affect and employee voice behaviour
Figure 1: conceptual model of key variables and hypotheses.
+ [H3.1], [H3.2] Active-constructive responses - [H4.1], [H4.2] Passive responses
Employee affect Employee voice behaviour Response to capitalization
+[H1] Positive affect - [H2] Negative affect
This conceptual model (figure 1) contains the following:
Hypothesis 1: Employee positive affect stimulates employee voice behaviour
Hypothesis 2: Employee negative affect inhibit employee voice behaviour
Hypothesis 3.1: Active-constructive responses to employee capitalization enhance the positive relationship between employee positive affect and employee voice behaviour
Hypothesis 3.2: Active-constructive responses to employee capitalization will positively influence the negative relationship between employee negative affect and employee voice behaviour
Hypothesis 4.1: Passive responses to employee capitalization will diminish the positive relationship between employee positive affect and employee voice behaviour
Hypothesis 4.2: Passive responses to employee capitalization will enhance the negative relationship between employee negative affect and employee voice behaviour
4. Method
4.1 Design
For this research a survey, more specifically a diary questionaire survey, is used in order to collect data needed to test the hypotheses. Data is collected in only one way, so it is a mono method study. We choose for a diary questionnaire survey because this method can study change and development over time (Saunders, Lewis & Thor, 2009). Respondents were asked to fill in a pre-test and to fill in three surveys a day for five consecutive working days long. A questionaires is the most suitable data
collection method for this research because it is for a bachor thesis. This means that time, finance and effort are not widely present. Questionaires are a quick and affordable way to reach a lot of respondents which are needed to provide reliable results (Saunders, et., 2009). Questionaires also provide standardised and consistent questions for all respondents which is needed in order to compare answers of different people and to make sure that comparison is realiable. The fact that it is a diary study is also what makes the study unique. Voice behaviour is not studied thoroughly in a longitudinal way. The questionaires are self-completed and web-based. Self-completed questionaires are less likely to be answered to please the researcher or because the respondent believes that certain responses are more socially desirable (Dillman, 2009). This makes the answers given by respondents more reliable.
To test the hypothesis described earlier we looked at variables on a certain, logical, moment in time. We used affect which was measured with the morning survey, capitalization which was measured with the lunch survey and employee voice behaviour which was measured during the afternoon survey. Due to this construction we can test whether affect in the morning and a response to capitalization during lunch will have an effect on employee voice behaviour during the afternoon.
4.2 Sample
All respondents in this study needed to be Dutch speeking full-time working employees in companies with more than two employees. They also needed to have a working internet connection and a device to fill in the questionnaires. It was tried to obtain an as large as possible sample size to make the results better generalizable (Saunders et al., 2009). The final sample consisted of 187 people who filled in at least one of the questionnaires. Most of the people in the sample fell in the age category 20-30 years old: 40.9%. This was followed by 18.9% in the age category of 51-60 years old, 17.7% in the category 41-50, 16.5% in the category 31-40, 3.7% in the category older than 60 and at last 2.4% in the category younger than 20 years old. The sample consisted for 39% of women and 61% of men. Of the respondents had 47.6% finished a HBO education followed by 31.1% who had finished
university, 16.5% who finished MBO and 4.9% who finished high school.
The final analysis were done over a sample of 65 respondents. This smaller sample is caused by people not filling in every survey needed to test the hypothesis described earlier. As mentioned earlier, we look at variables on a certain moment in time. When people did not fill in all needed surveys to test the hypothesis their data were deleted from the sample. This causes the diminishing sample from 187 respondents to only 65 respondents.
4.3 Measurements
4.3.1 Dependent variable: employee voice behaviour
The dependent variable, employee voice behaviour, was measured twice a day. The first time at lunch (around 12 pm) and the second time was in the afternoon (around 4 pm). At the two given points in time it was measured with a three item survey with a 7 likert scale. Those items were developed by Morrison, Liang, Fahr & Fahr and were tested on 500 respondents in a pre-study. An example item is: ‘This morning I spoke about ways to improve existing prcesses’. Scales range from (1) totally disagree to (7) totally agree. A high score means that an employee did voice about ways to improve existing processes. A low score means that an employee did absolutly not voice about ways to improve existing processes. Note that only Dutch speaking respondents were asked to fill in the questionnaires. The original scale was translated to Dutch.
4.3.2 Independent variable: employee affect
The independent variable, employee affect, was also measured twice a day. For the first time in the morning (around 9 am) and for the second time in the afternoon (around 4 pm). At the two given points in time it was measured with a 20 item survey with a 7 Likert scale. Those items were tested by the PANAS scale, developed by Watson, Clark and Tellegen in 1988. An example item is: ‘Today I
feel: stressed’. Scales range from (1) completely not to (7) completely. A high score means that the
employee’s mood at that moment is strongly related to the mood stated in the question. A low score 19
means that the employee’s mood is not related to the mood stated in the question. To measure this variable it was needed to translate the scale to Dutch. The Dutch translation however, does not have 10 items to measure positive affect and 10 items to measure negative affect. The Dutch translation of the PANAS scale uses 11 items to measure positive affect and 9 items to measure negative affect.
4.3.3 Moderator variable: response to capitalization
The moderator variable, response to capitalization, was also measured twice a day. The first time at lunch (around 12 pm) and the second time in the afternoon (around 4 pm). At the two given points in time it was measured with 1 item survey. Those items were tested by a scale developed by Gable, 2004. The item was: ‘What reaction did you get?’. The answer options where: positive and passive, positive and active, negative and passive, negative and active. Because this is not a scale variable but a categorical variable we transformed it into two dummy variables to be able to do regression analysis. To measure response to capitalization it was needed to translate the scale to Dutch.
4.4 Procedure
The questionnaires were created by our supervisor, Renske van Geffen. The program Qualtrics was used. This method saves time because the data did not have to be entered manually (Saunders et al., 2009). Also, the questionnaires did not need to be printed, which saved a lot of money due to the lack of printing costs (Wright, 2005). Collectively, a group of six students were all assigned to collect at least 30 respondents to fill in our surveys. Respondents were contacted through social media like Facebook and Twitter and were asked in person. By using social media it was easy to contact people in different geographical locations and in a very short period of time. For reaching my respondents I used my own network of friends and colleagues and I used my mother’s network. She is an
independent contractor for who networking is really important. Through her I reached people in all levels of organizations and all kinds of functions.
4.5 Analysis
This study examines a moderating effect. To find out whether the different responses to
capitalization alter the relationship between affect and voice behaviour two regressions models will be created. In the first regression model we will take a look at the separate effects of the
independent variables. The effect of employee positive and employee negative affect on the dependent variable employee voice behaviour. Also, the effect of moderating variable response to capitalization which consists of active-constructive responses and passive responses.
In the second regression model we will take a look at the interaction effects. We will look at the interaction effect of: active-constructive responses with positive affect, active-constructive responses with negative affect, passive responses with positive affect and passive responses with negative affect.
4.6 Predictions
In the first regression model we will look at the separate effects of the independent variables. We predict the following: A positive main effect of independent variable positive employee affect on the dependent variable employee voice behaviour; A negative main effect of independent variable negative employee affect on the dependent variable employee voice behaviour; A positive main effect of independent variable active-constructive response to capitalization on the dependent variable employee voice behaviour; A negative main effect of independent variable passive response to capitalization on the dependent variable employee voice behaviour
In the second regression model we will look at the interaction effects of the independent variables. We predict the following: A positive interaction effect of active-constructive responses to capitalization and positive employee affect on the dependent variable employee voice behaviour; A positive interaction effect of active-constructive responses to capitalization and negative employee affect on the dependent variable employee voice behaviour; A positive interaction effect of passive responses to capitalization and positive employee affect on the dependent variable employee voice
behaviour; A negative interaction effect of passive responses to capitalization and negative employee affect on the dependent variable employee voice behaviour.
5. Results
5.1 Reliabilities and correlations
To test whether the measurements mentioned earlier can be used for a regression analyses a reliability analysis is conducted. Table 1 shows the Cronbach’s alpha of the items on the diagonal. Because both types of responses to capitalization are dummies, no Cronbach’s alpha is given. Two of the three alphas are considered good while one is considered too low. The Cronbach’s alpha of employee voice behaviour is 0.551 but with deleting one item it would be 0.607. This item was not removed because employee voice behaviour is only measured with three items. Deleting one item more was not desirable. We choose to still use the employee voice behaviour scale in our analyses because the items were measured in a pre-test done by more than 500 respondents and this resulted in a Cronbach’s alpha of over 0.80. The employee positive affect scale was highly reliable (Cronbach’s alpha = .869). The employee negative affect was even more reliable (Cronbach’s alpha = .925). As expected there was a negative correlation between employee negative affect and
employee positive affect r(65) = -0.249, p<0.0. This was expected due to their opposed meaning. The correlation between active-constructive responses and passive responses to capitalization attempts was also expected to be negative due to their opposed meaning r(65) = -0.417, p<0.01. Also an expected correlation was that between employee positive affect and active-constructive responses
r(65) = 0.250, p<0.01. This correlation was expected due to the logical reasoning that people in a
more positive mood are more likely to produce an active-constructive response to a capitalization attempt. More unexpected is the finding that there is no correlation between employee voice behaviour and employee positive affect r(65) = 0.104, ns. A positive correlation was expected due to
the fact that positive affect leads to more organizational spontaneity and therefore more voice behaviour. The correlation between employee voice behaviour and employee negative affect was also expected to be negative due to opposed reasoning as described above r(65) = 0.08, ns. Also the correlations between employee voice behaviour and both the responses to capitalization were unexpected. A positive correlation was expected between employee voice behaviour and active-constructive responses r(65) = 0.016, ns, and a negative correlation was expected between employee voice behaviour and passive responses r(65) = 0.043, ns. Also unexpected was the fact that the affective states do not correlate with the responses to capitalization, besides employee positive affect and active-constructive responses. Furthermore, table 1 shows the correlation between variables and the Cronbach’s alpha on the diagonal.
Table 1. Descriptives and correlations between the variables (Cronbach’s Alphas on diagonal)
M SD 1 2 3 4 5 1.DV: Employee voice behaviour 4.9538 1.2640 (.551) 2. Employee positive affect 4.9281 .8979 .104 (.869) 3. Employee negative affect 1.9299 .8501 .080 -.249** (.925) 4. Active-constructive Responses .016 .250** -.085 5. Passive responses .043 -.059 .048 -.417** Note: N=65, *p<.05, **p<.01 5.2 Results
The first model shown in figure 2.1 and table 2, with an explained variance of 2.8 percent, shows the four main effects on employee voice behaviour. Only one variable had a sizable effect on employee voice behaviour, namely positive employee affect (β=.131, p<.05, R² = .028). This effect was
predicted and suggests that people in a positive mood are more likely to engage in employee voice behaviour.
The absent of the direct effect of negative employee affect was not predicted (β=.111, ns, R² = .028). This indicates that negative employee affect does not inhibit employee voice behaviour or has an effect at all. This suggests that when employees are in a negative mood that it won’t influence their voice behaviour. The direct effects of both the responses to capitalization were not predicted and were also not found. All direct effects can be seen in figure 2.1.
Figure 2.1: model 1 including all main effects. R² = .028
Active-constructive: β=.013, ns Passive: β=.051, ns
Positive: β=.131* Negative: β=.111, ns
The second model (figure 2.2 and table 2) shows all interaction effects. No single interaction effect was found to be significant. The interaction effect between employee positive affect and active-constructive responses was expected to enhance the already positive relationship between positive employee affect and employee voice behaviour. However, a negative interaction effect was found here (β= -.042, ns, R² = .036). The interaction effect between employee positive affect and passive responses was expected to diminish the positive relationship between employee positive affect and employee voice behaviour. However, a positive interaction effect was found here (β= .538,
ns, R² = .036). The interaction effect between employee negative affect and active-constructive
responses was expected to positively influence the negative relationship between employee negative
Employee affect Employee voice behaviour Response to capitalization
affect and employee voice behaviour. This interaction effect was, as expected, slightly positive (β= .022, ns, R² = .036). The last interaction effect, the effect between employee negative affect and passive responses, was expected to enhance the already negative relationship between employee negative affect and employee voice behaviour. This interaction was, as expected, negative (β= -.139,
ns, R² = .036).
The second model with interaction effects does predict slightly more variance than model 1 does. The second model explained 3.6 percent of the variance in comparison with 2.8 percent in model 1. This suggests that the four main effects in model 1 are less able to explain the dependent variable than the interaction variables in model 2. All interaction effects can be seen in figure 2.2.
Figure 2.2: model 2 including all interaction effects. R² = .036
Active-constructive β= -.042, ns Passive β= 0.538, ns
Positive affect
β=.022, ns β= -.139, ns
Negative affect
Employee affect Employee voice behaviour Response to capitalization
Table 2: Regression results of main model (model 1) and interaction effects (model 2) of positive
employee affect, negative employee affect, active-constructive and passive responses on employee voice behaviour.
Employee voice
behaviour (DV) Model 1 Model 2
Coefficient SE Beta Coefficient SE Beta
Constant 3,681*** 0,472 3,783*** 0,711 Employee positive affect 0,184* 0,083 0,131 0,154 0,122 0,109 Employee negative affect 0,165 0,085 0,111 0,185 0,137 0,125 Active-constructive responses 0,034 0,16 0,013 0,108 1,064 0,043 Passive responses 0,165 0,198 0,051 -1,15 1,372 -0,354
Employee positive affect *
active-constructive responses -0,02 0,18 -0,042
Employee positive affect * passive
responses 0,358 0,257 0,538
Employee negative affect *
active-constructive responses 0,026 0,191 0,022
Employee negative affect * passive
responses -0,203 0,232 -0,139
R² 0,028 0,036
Note: Dependent variable is employee voice behaviour, N=65 * p <.05,
** p<.01, *** p<.001
6. Discussion
6.1 Summary of results
The main goal of this study was to look at the effects of employee affect on employee voice behaviour. This is studied in a context with two kinds of response to capitalization attempts. The research question which had to be answered was: Do active-constructive and passive responses to employee capitalization influence the relationship between employee affect and employee voice behaviour?
Six hypotheses were stated and tested with the collected data. Hypothesis 1 stated that employee positive affect would have a positive influence on employee voice behaviour. This hypothesis was confirmed during this study. This finding puts forward what literature already suggested: positive moods have a positive influence on the levels of organizational citizenship behaviour. Those behaviours include voicing. The confirmation of this hypothesis holds that when employees are in a positive mood they are more likely to produce voice behaviour.
Hypothesis 2 stated that employee negative effect would have a negative influence on employee voice behaviour. The relationship between negative moods and voice behaviour, or even organizational citizenship behaviour, was never studied before. The negative effect was not found and suggests therefore that when employees are in a negative mood, it won’t influence their voice behaviour.
Hypothesis 3.1 stated that when employees experience a positive mood and they get an active-constructive response to their capitalization response they would produce even more voice behaviour than when they would only experience a positive mood. This hypothesis was not
confirmed which suggests that an active-constructive response won’t lead to more voice behaviour when employees are in a positive mood. This result was unexpected because theory suggests that when people get active-constructive responses to their capitalization attempts, they are more likely to be in an even more positive mood than when they did not capitalize. Because the hypothesized
positive relationship between employee positive affect and employee voice behaviour was confirmed, it would be more logical that this hypothesis would also be confirmed. Alternative explanations will be discussed later.
Hypothesis 3.2 stated that when employees experience a negative mood and they get an active-constructive response to their capitalization response they would produce even less voice behaviour than when they would only experience a negative mood. This hypothesis is mostly based on hypothesis 2 which suggests that there is a negative relationship between employee negative affect and employee voice behaviour. However, hypothesis 2 was not confirmed which makes hypothesis 3.2 also less logical. The fact that hypothesis 3.2 is not confirmed suggests that an active-constructive response to a capitalization attempt won’t lead to more voice behaviour when
employees are in a negative mood.
Hypothesis 4.1 stated that when employees experience a positive mood and they get a passive response to their capitalization attempt, they are likely to produce voice behaviour, but less voice behaviour than when they did not capitalize. This hypothesis was not confirmed which suggests that a passive response won’t lead to less voice behaviour when employees are in a positive mood. This finding was unexpected due to outcomes in literature which suggested that passive responses to capitalization attempts were correlated with higher negative affect.
Finally, hypothesis 4.2 stated that when employees experience a negative mood and they get a passive response to their capitalization attempt, they are likely to produce even less voice
behaviour than when they did not capitalize. This hypothesis was also not confirmed. This suggests that even when employees experience a negative mood state, their voice behaviour will not be influenced by a passive response to their capitalization attempt. This was a logical consequence of the fact that hypothesis 2 was also not confirmed. The fact that hypothesis 2 and hypothesis 4.2 were also not confirmed suggests that there is no relation between negative affect and voice behaviour.
6.2 Unexpected results and alternative explanations
The fact that only one of the six hypotheses is confirmed in this study suggests that there is a possibility of alternative explanations. In this section the alternative explanations will be sought. First of all, hypotheses 3.1, 3.2, 4.1 and 4.2 are all based on the first two hypotheses. The fact that only hypothesis 1 is confirmed makes it very hard for hypotheses 3.2 and hypothesis 4.2 to be confirmed because they are based on a relationship stated in hypothesis 2 which was not confirmed.
Second, the relationship between affect and voice behaviour was never studied as a direct relationship before. George and Brief (1992), Bateman and Organ (1983) and Organ and Ryan (1995) never looked at the influences of negative affect on behaviours. Only Williams and Shiaw (1999) stated that positive moods are more likely to have a positive impact on organizational citizenship behaviour than negative moods. This study assumed that when employee positive affect would have a positive influence on employee voice behaviour, employee negative affect would have a negative influence on employee voice behaviour. This study suggests that this assumption earlier made is clearly not the case. This study did not only focus on affect and voice behaviour but had also lots of other variables were present in the surveys. This resulted in only three items focussing on voice behaviour and nine items focussing on negative affect. Although the relation between negative affect and employee voice behaviour was not confirmed in this study, we still suggest that it could be possible. This leaves room for future studies to do more detailed research on this specific relation and perhaps on other relationships between negative affect and behaviours.
Third, this study assumed that affect influenced organizational citizenship behaviour and therefore voice behaviour. It is possible that the influences of affect on organizational citizenship behaviour are on different aspects of organizational citizenship behaviour than on, more specifically, voice behaviour. This study measured specifically employee voice behaviour and not organizational citizenship behaviour. This could explain not confirming the hypotheses because the relationship on which those hypotheses were based, was not there.
Fourth, all hypotheses considering responses to capitalization attempts were not confirmed. 29
The hypotheses created in the theoretical framework were based on literature stating that active-constructive responses to capitalization attempts are positively correlated to positive affect,
satisfaction, commitment, intimacy and trust. Correlation suggests a relation between two variables but does not state a causal effect. The lack of interaction effects found in the hypotheses about responses to capitalization attempts could suggest that those responses do not have the calculated impact on affect and therefore on voice behaviour. This holds that it is possible that
active-constructive responses did not gave positive affect a boost and therefore no positive boost on employee voice behaviour. The same reasoning can be applied for negative affect and the positive influence that active-constructive was hypothesized to have on the negative relationship between negative employee affect and voice behaviour.
Passive responses are, as described in the theoretical framework, negatively correlated to affect, satisfaction, commitment, intimacy and trust. The fact that the hypotheses considering passive responses to capitalization attempts were also not confirmed could be due to the same reasoning described above. Also, here holds that correlation does not state a causal effect. The lack of interaction effects between positive affect and passive responses and negative affect and passive responses could suggest that those responses do not have the considered impact on affect and therefore on voice behaviour. For the hypotheses considering passive responses holds that it is possible that the passive responses did not inhibit positive affect or enlarge negative affect and therefore decrease voice behaviour.
Finally, this study suggests a causal relationship between affect and voice behaviour. What also could be true is that this relationship is reversed so that voice behaviour influences affect. It could be possible that by making constructive suggestions that employees feel more useful or happy which could influence their affect.
6.3 Discussion points
This study focussed on influences of employee voice behaviour. Despite that employee voice behaviour is the most important variable, it was only measured with three items. This resulted in a low Cronbach’s alpha. We still chose to use this variable because the pre-test, which was tested under 500 respondents, resulted in a very reliable score (>.80). Using a variable with a Cronbach’s alpha of .551 means that 69.64% of the test score is attributable to random error (Nunnally & Bernstein, 1967). This suggests that when this exact same study will be done again, whole different outcomes can be found due to the high percentage which can be explained by random error. What made this study relevant, useful and unique was the way data was collected. The data was collected through means of a diary study. This diary study gave us the opportunity to better assign the causal effects found during this study. This due to the fact that employee affect was measured in the morning, responses to capitalization were measured during lunch and employee voice behaviour was measured in the afternoon. This kind of study, were employees fill in three surveys a day, is unique and will contribute to the literature about voice behaviour.
The diary study was also one of the weakest points of this study. Due to the nature of work nowadays, a lot of surveys had not been filled in. It was asked of employees to fill in three surveys a day for a whole work week long. The problem with the design lies in the fact that work nowadays is not structured like the surveys were structured. Lots of people do not work five days a week from nine till five. This resulted in a very low response rate. Only 187 people managed to fill in at least one whole survey. For the analysis done in this study this resulted in analysis done over data collected from 65 respondents.
6.4 Contributions, practical implications and future research
As stated earlier, this study contributes in two ways. First it contributes to the already extensive literature about voice behaviour of employees. This study showed that positive affect experienced by employees influenced voice behaviour expressed by employees. This has also practical relevance
since managers in companies should try to influence the moods of employees to make them voice more. As stated in literature before by Williams and Shiaw (1999), managers could focus on things like the physical environment in which employees work. Think about all factors which could influence the mood of employees. Factors like temperature, workspace, ambiance and other factors could influence employee mood and therefore employee voice behaviour.
The fact that negative moods of employees do not influence their voice behaviour has also practical implications for managers. This finding suggests that managers won’t have to worry about employees experiencing negative moods when it comes to their voice behaviour.
Responses to capitalization attempts done by employees do not have an effect on the voice behaviours expressed by employees despite which mood they experience. This has practical
implications for managers who have to manage larger groups of employees who are likely to influence each other on the job. The lack of effect of response to capitalization attempts suggests that managers won’t have to worry about how employees react to each other’s capitalization efforts. For managers it would have been hard to influence the way employees respond to others
capitalization attempts. Thanks to this study this is one less thing managers have to worry about. Future studies could elaborate the findings done in this study. This study found that positive moods have a positive impact on employee voice behaviour. With this in mind, future research could find out which specific moods positively influence voice behaviour. This would also generate even more practical implications for managers. When future findings suggest that when people are highly interested (positive mood), they are more likely to voice. Those are findings which could really make a difference in companies because managers could try to stimulate employees to be interested in one another.
Future studies could also try to replicate this study but only with a less strict period of time in which surveys have to be filled in. Due to the modern working conditions in which almost nobody works five days a week from nine till five, this study generated a low response rate. Future studies should learn from this study and could, for example, ask employees to fill in three surveys a day for
five days which do not have to be in one week. This will lead to a higher response rate which will make the analysis and results more trustworthy and generalizable.
6.5 Conclusion
This study was designed to find out whether active-constructive and passive responses to capitalization attempts changed the relationship between employee affect and employee voice behaviour. The findings of this study suggest that both kinds of responses have no influence on the relationship between employee affect and employee voice behaviour. The relationship between employee positive affect and employee voice behaviour has been confirmed whereas the
relationship between employee negative affect and employee voice behaviour has not been found. To come back to the example stated in the introduction in which being omniscient in the business world was identified as impossible but being well informed was called crucial. This study suggests that making sure that employees talk constructively about ideas and make useful suggestions can be influenced by making sure that employees experience a positive mood. This positive mood can be influenced by managers to make sure that employees voice as much as possible.
References
Ashford, S. J., Rothbard, N. P., Piderit, S. K., & Dutton, J. E. (1998). Out on a limb: The role of context
and impression management in selling gender-equity issues. Administrative Science
Quarterly, 23-57.
Ashford, S. J., Sutcliffe, K. M., & Christianson, M. K. (2009). Speaking up and speaking out: The
leadership dynamics of voice in organizations. Voice and silence in organizations, 175-202.
Athanassiades, J. C. (1973). The distortion of upward communication in hierarchical
organizations. Academy of Management Journal, 16(2), 207-226.
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.
Baron, R. A. (2008). The role of affect in the entrepreneurial process. Academy of Management
Review, 33(2), 328-340.
Bateman, T. S., & Organ, D. W. (1983). Job satisfaction and the good soldier: The relationship
between affect and employee “citizenship”. Academy of management Journal, 26(4), 587-
595.
Block, M., & Zautra, A. (1981). Satisfaction and distress in a community: A test of the effects of life
events. American Journal of Community Psychology, 9(2), 165-180.
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.
Dillman, D. A., Smyth, J. D., & Christian, L. M. (2009). Internet, mail, and mixed-mode surveys: The
tailored design method . Hoboken.
Frijda, N. H. (1993). De emoties: Een overzicht van onderzoek en theorie. Bakker.
Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do you do when things go right? The
intrapersonal and interpersonal benefits of sharing positive events. Journal of personality and
social psychology, 87(2), 228.
George, J. M. (1991). State or trait: Effects of positive mood on prosocial behaviors at work. Journal
of applied Psychology, 76(2), 299.
George, J. M. (1992). The role of personality in organizational life: Issues and evidence. Journal of
Management, 18(2), 185-213.
George, J. M., & Brief, A. P. (1992). Feeling good-doing good: a conceptual analysis of the mood at
work-organizational spontaneity relationship.Psychological bulletin, 112(2), 310.
Glauser, M. J. (1984). Upward information flow in organizations: Review and conceptual
analysis. Human Relations, 37(8), 613-643.
Graham, J. W. (1986, August). Organizational citizenship informed by political theory. In annual
meeting of the Academy of Management, Chicago, IL.
Grant, A. M., & Ashford, S. J. (2008). The dynamics of proactivity at work.Research in organizational
behavior, 28, 3-34.
Howard, A. (1995). A framework for work change. The changing nature of work,3, 44. Higgins, E. T. (1997). Beyond pleasure and pain. American psychologist,52(12), 1280.
Isen, A. M., Shalker, T. E., Clark, M., & Karp, L. (1978). Affect, accessibility of material in memory, and
behavior: A cognitive loop?. Journal of personality and social psychology, 36(1), 1.
Isen, A. (1987). Positive affect, cognitive processes, and social behavior.Advances in experimental
social psychology, 20, 203-253.
Katz, D. (1964). The motivational basis of organizational behavior. Behavioral science, 9(2), 131-146. 35
Katz, D., & Kahn, R. L. (1978). The social psychology of organizations.
Langston, C. A. (1994). Capitalizing on and coping with daily-life events: Expressive responses to
positive events. Journal of Personality and Social Psychology, 67(6), 1112.
Lewinsohn, P. M., & Graf, M. (1973). Pleasant activities and depression.Journal of consulting and
clinical psychology, 41(2), 261.
Lewinsohn, P. M., & Libet, J. (1972). Pleasant events, activity schedules, and depressions. Journal of
abnormal psychology, 79(3), 291.
Liang, J., Fahr, C., & Fahr, L. (2012). A two-wave examination of the psychological antecedents of
voice behavior. Academy of Management Journal,55, 71-92.
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(1), 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(6), 1453-1476.
Morrison, E. W. (2011). Employee voice behavior: Integration and directions for future research. The
Academy of Management Annals, 5(1), 373-412.
Morrison, E. W., & Milliken, F. J. (2003). Speaking up, remaining silent: The dynamics of voice and
silence in organizations. Journal of Management Studies, 40(6), 1353-1358.
Nunnally, J. C., Bernstein, I. H., & Berge, J. M. T. (1967). Psychometric theory(Vol. 226). New York: McGraw-Hill.
Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Lexington
Books/DC Heath and Com.
Organ, D. W., & Ryan, K. (1995). A meta-analytic review of attitudinal and dispositional predictors of
organizational citizenship behavior. Personnel psychology, 48(4), 775-802.
Parrott, W. (2001). Emotions in social psychology: Essential readings. Psychology Press.
Pinder, C. C., & Harlos, K. P. (2001). Employee silence: quiescence and acquiescence as responses to
perceived injustice. Research in personnel and human resources management, 20, 331-369.
Premeaux, S. F., & Bedeian, A. G. (2003). Breaking the Silence: The Moderating Effects of Self-
Monitoring in Predicting Speaking Up in the Workplace*. Journal of Management
Studies, 40(6), 1537-1562.
Rusbult, C. E., Zembrodt, I. M., & Gunn, L. K. (1982). Exit, voice, loyalty, and neglect: Responses to
dissatisfaction in romantic involvements. Journal of Personality and Social Psychology, 43(6),
1230.
Saunders, M., Lewis, P., & Thornhill, A. (2009). Research methods for business students. Essex: Prentice Hall.
Tangirala, S., & Ramanujam, R. (2008). Exploring nonlinearity in employee voice: The effects of
personal control and organizational identification. Academy of Management Journal, 51(6),
1189-1203.
Taylor, S. E. (1991). Asymmetrical effects of positive and negative events: the mobilization
minimization hypothesis. Psychological bulletin, 110(1), 67.
Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-role behaviors: Evidence of construct and
predictive validity. Academy of Management Journal,41(1), 108-119.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of
positive and negative affect: the PANAS scales. Journal of personality and social
psychology, 54(6), 1063.
Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological
bulletin, 98(2), 219.
Williams, S., & Shiaw, W. T. (1999). Mood and organizational citizenship behavior: The effects of
positive affect on employee organizational citizenship behavior intentions. The Journal of
Psychology, 133(6), 656-668.
Wright, K. B. (2005). Researching Internet-based populations: Advantages and
disadvantages of online survey research, online questionnaire authoring software
packages, and web survey services. Journal of Computer-Mediated Communication, 10(3).
