• No results found

The influence of negative work events on proactive behavior : a diary study on three antecedents of proactive behavior

N/A
N/A
Protected

Academic year: 2021

Share "The influence of negative work events on proactive behavior : a diary study on three antecedents of proactive behavior"

Copied!
78
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

 

     

The influence of negative work events on proactive

behavior:

A diary study on three antecedents of proactive behavior

MSc. Business Studies – Leadership & Management Track

Supervisor: Dr. Antje Schmitt

Second reader: Dr. Frank Belschak

Student: Quinyne Albertzoon – 5947154 Date: August 15, 2014

(2)

  2  

Table of contents

Abstract ... 3 1. Introduction ... 4 2. Literature review ... 7 2.1 Proactive behavior ... 7

2.2 Negative work events ... 9

2.3 Affect ... 10

2.4 Proactive motivational states ... 14

3. Research methods ... 18 3.1 Method ... 18 3.2 Participants ... 19 3.4 Measures ... 20 4. Results ... 26 5. Discussion ... 37

6. Limitations and future research directions ... 41

7. Conclusion ... 45

8. Bibliography ... 47

Appendix I – Invitation letter participants ... 49

Appendix II – Instruction letter participants ... 51

Appendix III – Question list questionnaires ... 53

Tables and figures

Figure 1 The Conceptual Model…….………6

Table 1 Overview of the Measured Variables...………..25

Table 2 Internal Reliability of Constructed Variables………...26

Table 3Correlations, Means, and Standard Deviations……….29

Table 4 Regression Analysis for Mediation of Positive and Negative Affect Regulation………...33

Table 5 Regression Analysis for Moderation of Role Breadth Self-Efficacy and Vitality……….36

(3)

         

  3  

Abstract

This study extends research on proactive behavior at work by examining the extent to which negative work events, positive and negative affect regulation, and the “can do” (role breadth self-efficacy) and “energized to” (vitality) motivational states are associated with proactive behaviors (voice and taking charge). A diary study is conducted among 105 participants, who are working in various sectors for at least 32 hours a week. Data on negative work events, positive and negative affect, role breadth self-efficacy, vitality, and proactive behavior is collected three times a day over 7 workdays. On the first day of the diary study a general questionnaire is filled out. From day 2 to 8 the daily questionnaires were completed. The data is analyzed from a between-persons perspective with the use of the Process tool for SPSS. The results showed that negative work events were not significantly related to proactive behaviors, positive and negative affect, and the two motivational states. A mediation effect by positive and negative affect regulation, and moderation effect by role breadth self-efficacy and vitality on the relationship between negative work events and proactive behavior was statistically not supported. Finally, the priorities for future research are summarized and a conclusion is given.

(4)

  4  

1. Introduction

In the current dynamic work environment it becomes increasingly important that employees show initiative and behave proactively at work. This is due to the continuous process of constant decentralization, fast-paced change, the increasing demand for innovation, and high operational uncertainty in organizations (Bindl & Parker, 2010; Crant, 2000; Fritz & Sonnentag, 2009). It is for this reason that organizations are in need of flexible and responsive employees who are able to approach work in a proactive manner by showing initiative without the need for constant supervision (Belschak & Hartog, 2010; Crant, 2000). Correspondingly, employees are less dependent on a single organization to build on their personal and career development (Belschak & Hartog, 2010; Bindl & Parker, 2010; Hartog & Belschak, 2007). Therefore, employees should take charge of their own personal development and career paths.

Following Parker, Bindl, and Strauss (2010) a general definition is introduced “proactive behavior is about making things happen, anticipating and preventing problems, and seizing opportunities” (p. 827). In order to reach the level of proactivity several antecedents can emerge, like positive or negative events, differing motivational states, and/or a positive or negative affective state (Bledow, Schmitt, Frese, & Kühnel, 2011; Bledow, Rosing, & Frese, 2013; Parker et al., 2010). Negative events are a form of affective events, because they have an influence on people’s mood (Bledow et al., 2011). According to the affective events theory, events are important incidents that create a change in circumstances and to which people react emotionally or change their mood (Bledow et al., 2011; Weiss & Cropanzano, 1996). The internal affect regulation process can regulate an individual’s emotional reaction to or mood after experiencing a negative event (Bledow et al., 2011). The “energized

(5)

         

  5  

to” and “can do” motivational states can influence people’s behavior (Parker et al., 2010). Energized to “refers to activated positive affective states that prompt proactive goal processes” (Parker et al., 2010, p. 827). The can do motivational state “arises from the perceptions of self-efficacy, control, and (low) cost” (Parker et al., 2010, p. 827).

The aim of this thesis is to explore how affect regulation mediates the relationship between negative work events and employees’ daily proactive behavior. In addition, the moderator effect of the two motivational states, “energized to” and “can do”, will be assessed on the negative work events – employees’ daily proactive behavior link. According to Tornau and Frese (2013) scholars have increasingly gained interest in proactivity research, which caused that more studies have been conducted on the antecedents and consequences of specific forms of proactivity. However, previous research has not been able to reach an integrated theoretical framework to gain more understanding of the relationships between the proactive constructs and its antecedents and outcomes (i.e. the nomological net) (Crant, 2000; Grant & Ashford, 2008; Tornau & Frese, 2013). Let alone investigating the mediating role of affect regulation in the relationship between negative work events and daily proactive behavior. Also, scholars have not combined this relationship with the “energized to” and “can do” motivational states moderating on the negative work events and daily proactive behavior link. Therefore, the research questions addressed in this thesis are:

“How does affect regulation mediate the relationship between negative work events and daily proactive work behavior?” and “How do the “energized to” and “can do” motivational states moderate the negative work events and daily proactive behavior relationship?”

(6)

  6   In order to answer the research questions a diary study of eight workdays will be conducted. The literature review, the following chapter, elaborates on proactive behavior, negative work events, positive and negative affect, and the can do and energized to motivational states. In this chapter the hypotheses to test my assumptions are stated. In order to gain a clear overview of the relationships, which are tested in this study, see Figure 1. Subsequently, the research method will be explained followed by the results. Third, the discussion on the findings is given, followed by the limitations and recommendations for future research. Finally, the conclusion is stated.

Figure  1  The  conceptual  model

     

(7)

 

2. Literature review

The introduction stated that proactive behavior is highly important in the current dynamic organizational field. Gaining further knowledge on its antecedents, e.g. negative work events, affect, and motivational states, is key in order to develop theories and strategies to enhance employees’ proactivity. This chapter provides an in-depth elaboration on important variables of this study and the hypotheses are stated.

2.1 Proactive behavior

In order to understand what proactivity signifies the following definition clarifies what is meant with proactive behavior in this study. Proactive behavior is defined as “self-directed and future-focused action in an organization, in which the individual aims to bring about change, including change to the situation (e.g. introducing new work methods, influencing organizational strategy) and/or change within oneself (e.g. learning new skills to cope with future demands)” (Bindl & Parker, 2010, pp. 3-4). Proactive behavior is about taking matters in to own hands, anticipating, and preventing problems to enhance the current situation or to create new opportunities (Crant, 2000; Parker et al., 2010) In the end, proactive behavior contains three key attributes; it is self-starting, change oriented, and future focused (Parker et al., 2010; Tornau & Frese, 2013).

Recently the interest in proactive work behavior has increased in the academic and organizational field (Bindl & Parker, 2010; Parker et al., 2010). Scholars have tried to capture the developments of proactive concepts, because there are currently different views on whether proactivity is a stable disposition, a pattern of behaviors, or a way of behaving at work (Bindl & Parker, 2010; Crant, 2000; Den Hartog &

(8)

Belschak, 2007; Grant & Ashford, 2008; Tornau & Frese, 2013). In this thesis proactivity is conceptualized as a way of behaving at work.

Proactive behavior is measured by different variables, such as voice, taking charge, and proactive personality (Den Hartog & Belschak, 2007; Belschak & Den Hartog, 2010; Tornau & Frese, 2013). In this research the variables voice and taking charge are used to measure proactive behavior. These two constructs are chosen, since they are common used key attributes of proactivity and are likely to occur in various organizational contexts (Tornau & Frese, 2013). Voice and taking charge refer to specific behaviors in contrast to proactive personality (Tornau & Frese, 2013). The latter is characterized as a stable disposition (Bindl & Parker, 2010; Tornau & Frese, 2013); therefore, it functions in this study as a control variable.

Voice is a challenging promotive behavior and can be defined as constructive change-oriented communication to improve the status quo (Van Dyne & Lepine, 1998; LePine & Van Dyne, 2001). Challenging promotive behavior entails that a person signals misconduct and acts up on it by exposing the misconduct to others, but with the intention to improve the situation (Van Dyne & LePine, 1998; Tornau & Frese, 2013). In the end, voice is a proactive behavior that is intended to communicate misconduct and to propose ideas to improve the current situation no matter other opinions.

Taking charge can be defined as efforts that are undertaken voluntarily and on a constructive basis by individual employees, to bring about organizational change, with respect to the execution of work within the different layers of the organization’s context (e.g. their job, teams, departments, or organizations) (Morrison & Phelps, 1999). The taking charge construct is designated as being discretionary (i.e. it is not

(9)

 

formally required), inherently change-oriented and aimed at improvement (Fritz & Sonnentag, Antecedents of day-level proactive behavior: A look at job stressors and positive affect during the workday, 2009; Morrison & Phelps, 1999; Tornau & Frese, 2013). In addition, similar to voice researchers refer to taking charge as challenging promotive behavior (Morrison & Phelps, 1999).

2.2 Negative work events

Negative work events might be experience in their daily course of action by an individual. Here events are defined as “significant happenings that produce a change in circumstances” (Bledow et al., 2011, p. 1248) and “generate an emotional reaction or mood change in people” (Weiss & Cropanzano, 1996, p. 31). Events are, according to the affective events theory (Weiss & Cropanzano, 1996), situational antecedents of affect. The reason to define events as situational antecedents of affect arises from the idea that events influence the affective state of an individual (Bledow et al., 2011; Ohly & Fritz, 2007; Weiss & Cropanzano, 1996).

A distinction is made between positive and negative work events (Bledow et al., 2011; Bledow et al., 2013; Fritz & Sonnentag, 2009; Ohly & Schmitt, 2013). Positive events are in line with achieving goals and do not stand in the way of work tasks. An example of a positive event is receiving positive feedback, which in turn causes the affect of the receiver to increase positively. Negative events, on the other hand, are in this study seen as situational constraints (e.g. regulation obstacles or work barriers), which distract employees from their work tasks (Fritz & Sonnentag, 2009). Negative work events at first deregulate proactive behavior, because when an individual has to deal with them there might be no time or cognitive resources to behave proactively (Bledow et al., 2011; Fritz & Sonnentag, 2009). On the long run

(10)

when an individual experiences negative work events he/she might be able to engage in proactive behaviors due to an inconsistent image between the desired and actual situation (Fritz & Sonnentag, 2009; Parket et al., 2010). The affective events theory proposed by Weiss and Cropanzano (1996) concludes that the occurrence of both positive and negative events play a role in pursuing work-related goals.

In this research it is assumed that negative events have a negative influence on proactive behavior (voice and taking charge), thus there is a negative relationship. Therefore, the following hypotheses are derived;

Hypothesis 1a: Negative work events will be negatively associated with voice.

Hypothesis 1b: Negative work events will be negatively associated with taking charge.

2.3 Affect

In this study affect is “defined as an individual's ‘direct emotional response’ and ‘immediate and certain emotion’ connected with the thought of engaging in the behavior” (Triandis, 1982 found in Lauver, 1992). This definition explains that affect is the emotional response connected with the initiative to engage in some type of behavior. This research is about the connection between affect and behavior, therefore, this definition is used.

Affect is divided into two dimensions of the same continuum (Bledow et al., 2011), namely positive and negative affect. Bledow et al. (2011) and Bledow et al. (2013) mention that positive and negative affect can be experienced within a time interval. However, it is rare to experience both affective states simultaneously, especially when the levels of the affective states are both high (Bledow et al., 2011; Bledow et al., 2013). The personality systems interaction (PSI) theory by Kuhl (2000) explains how these affective states are intertwined.

(11)

 

The PSI theory integrates the idea of how positive and negative affect are regulated and the theory distinguishes itself by the focus on change in affect. Kuhl (2000) differentiates for both affective states two levels of affect, which people can experience, namely low or high. Low positive affect is perceived to regulate cognition in a controlled, slow, and sequential mode (Bledow et al., 2013). High positive affect, on the contrary, enables the regulation of cognition in an automatic, fast, and parallel manner. High negative affect regulates “whether attention is narrow and focused on isolated elements” (Bledow et al., 2013, p. 433). In contrast, low negative affect regulates whether attention is “broad and inclusive of the context” (Bledow et al., 2013, p. 433). When in the following chronological order negative affect is followed by positive affect an affective shift occurs (Bledow et al., 2011). Thus, affective shifts are the decrease in negative affect followed by the increase of positive affect.

Based on the PSI theory Bledow et al. (2011) have developed the affective shift model. The affective shift model (Bledow et al., 2011) proposes that the experience of negative affect followed by positive affect leads to work engagement. The model is based on two mechanisms, namely the down-regulation of negative affect and the up-regulation of positive affect, which could be referred to as affect regulation (Bledow et al., 2011). An individual effectively regulates affect by being able to implicitly and intuitively down-regulate negative affect and to up-regulate positive affect after negative events have occurred. The occurrence of negative work events interrupts the effective affect regulation process. Therefore, it is expected based on the affective shift model, when a negative event occurs at first positive and negative affect will be ineffectively regulated. This means that positive affect will be down regulated and negative affect will be up regulated short after the experience of a negative work event. The following hypotheses are proposed:

(12)

Hypothesis 2a: Negative work events will be negatively associated with the up-regulation of positive affect.

Hypothesis 2b: Negative work events will be negatively associated with the down-regulation of negative affect.

According to Den Hartog and Belschak (2007) researchers have acknowledged that proactive behaviors are driven by the affective system of an individual. Scholars have tried to gain insight on how affect can influence proactive behaviors (Fritz & Sonnentag, 2009; Bledow et al., 2013), meanwhile, e.g. Fritz and Sonnentag (2009) focused only on how positive mood effected proactive behavior and they did not include the effect of negative mood. The choice to solely include positive mood was among others based on the assumption of its relevance and the research by Frederickson (2001). Frederickson’s (2001) broaden-and-build model proposes that positive mood broadens the thought-action repertoire. This means that individuals are open to develop their personal resources, which can range from physical and intellectual resources to social and psychological resources (Frederickson, 2001). In addition, self-regulation theories emphasize that positive affect influences individuals to set high goals for tasks and to undertake action to achieve those goals (Bledow et al., 2011). Even though, negative affect is necessary to experience a positive behavioral change (Bledow et al., 2011; Bledow et al., 2013; Frederickson, 2001). The initiation of negative affect increases the generation of new ideas due to the high level of persistence held towards a task (De Dreu, Baas, & Nijstad, 2008). The effect of negative affect can be twofold (Den Hartog & Belschak, 2007). On the one hand, negative affect can trigger the coping mechanisms that cause the resources of individuals to decrease, thus taking their availability away from other tasks (Den Hartog & Belschak, 2007, p. 606). Meaning that negative affect disturbs the current

(13)

 

ongoing actions, which causes a person’s mental processes and rapid mobilization to be interrupted (Bledow et al., 2011). This might imply that behaviors are affected in a negative manner. On the other hand, it is also been said that negative affect can stimulate an active attitude, rather than a negative one (Den Hartog & Belschak, 2007, p. 606). Negative affect possesses a key self-regulatory function, which can be the foundation to commit oneself to a task at a later point in time (Bledow et al., 2011). Moreover, negative affect is known “[to facilitate] an analytic mode of information processing in which people analyze information step by step, pay close attention to details, and are sensitive to discrepancies” (Bledow et al., 2011, p. 1247). This mode of information processing causes that people gain a thorough and realistic view of the situation, and prepare for taking goal-directed action. Thus, negative affect can create an urgency among people that things are not going as planned and action needs to be taken. This shows that under certain conditions negative affect can be of motivational nature. As a result of the pros and cons of positive and negative affect, one can conclude that both positive and negative affect, especially the interplay between the two have critical functions in the relationship with behavior.

The affective experiences of individuals change continuously, external events cause emotions to rise and fall, and moods are subject to ongoing gradual change (Bledow et al., 2011; Bledow et al., 2013; Fritz & Sonnentag, 2009; Parker et al., 2010). It is assumed that effective positive affect regulation causes an individual to initiate the behavior of expressing voice and taking charge. This assumption is based on the idea that positive affect enables individuals to initiate change-oriented communication and propose opportunities in a constructive manner when experiencing misconduct (Bledow et al., 2011; Bledow et al., 2013; Fritz & Sonnentag, 2009; Parker et al., 2010). For negative affect regulation it is expected that

(14)

when it is ineffectively regulated that engaging in proactive behaviors will not be the case. This expectation is based on the notion that negative affect is only effective in future states when it is down-regulated and on the short run it interrupts the engagement in current actions, which causes cognitive processes to be interupted too (Bindl & Parker, 2010; Bledow et al., 2011 Bledow et al., 2013). The following hypotheses are derived;

Hypothesis 3a: Positive affect regulation will be positively related to voice.

Hypothesis 3b: Positive affect regulation will be positively related to taking charge.

Hypothesis 3c: Negative affect regulation will be negatively related to voice.

Hypothesis 3d: Negative affect regulation will be negatively related to taking charge.

Hypothesis 4: Positive and negative affect regulation will mediate the relationship between negative work events and daily proactive behavior.

2.4 Proactive motivational states

Individuals have different reasons for pursuing proactive goals and preferring one to another, also do the goals differ from person to person (Parker et al., 2010). In order to understand the reasons behind proactive behavior the personality traits are not sufficient as the sole motivator of proactive behavior. It is necessary to gain an understanding of the individual’s motivational state in the corresponding context and in relation to the envisioned future (Parker et al., 2010). Parker et al. (2010) proposed three motivational states, namely “can do”, “reason to”, and “energized to”. In this study the focus is on two motivational states, namely can do and energized to.

The can do motivational state is about the perception of one owns ability, control appraisals and attributions, and the costs involved of taking action (Parker et

(15)

 

al., 2010). In this state the individual asks his/herself if he/she can perform a certain task and assesses the likely outcomes of his or her behaviors when setting a proactive goal. The thought that an individual can be successful in a particular domain, or high self-efficacy, is seen as important in proactive goal generation because there is a high potential psychological risk involved in being proactive. By using one’s personal initiative and taking charge to improve work methods the people in the environment can react by resisting the change and being skeptic. The same accounts for active feedback seeking, which can cause damage to the person’s ego and perceived image. Therefore, it is necessary for individuals to feel confident to start proactive goals and deal with their consequences before they act (Parker et al., 2010, p. 834).

The can do motivational state can be measured by different concepts, two examples are, self-efficacy and role breadth self-efficacy (Ohly & Fritz, 2007; Parker, 1998; Tornau & Frese, 2013). Self-efficacy, refers to the ability of the self to organize and perfom tasks that are prescribed in ones job role (Ohly & Fritz, 2007; Parker, 1998; Tornau & Frese, 2013). On the other hand, Parker (1998) expresses role breadth self-efficacy as “employees’ perceived capability of carrying out a broader and more proactive set of work tasks that extend beyond prescribed technical requirements” (p. 835). Moreover, role breadth self-efficacy relates to the extent to which people feel like they can successfully complete broader and more proactive tasks, beyond their traditional job description (Ohly & Fritz, 2007; Parker, 1998; Tornau & Frese, 2013). Tornau and Frese (2013) explain that role breadth self-efficacy is an adaptable state and an antecedent of proactive behavior. Ohly and Fritz (2007) found that role breadth self-efficacy is related to proactive behavior and recommend its use in future research since it targets behavior beyond what is formally required in a given job. Therefore, I

(16)

decided to measure the can do motivational state with the concept of role breadth self-efficacy.

However, researchers have not paid much attention to the moderating effects of role breadth self-efficacy on the relationship between negative work events and practive behavior. It is assumed that when negative work events occur the act of performing proactive behavior will be low, especially when the can do motivational state is low. On the contrary, it is expected when negative events occur and the motivational state is high that proactive behavior can still be high. Therefore, the following two hypotheses will test whether the assumptions are correct;

Hypothesis 5a: Negative work events will be negatively related to voice when an individual’s motivational state, can do, is low and not when it is high.

Hypothesis 5b: Negative work events will be negatively related to taking charge when an individual’s motivational state, can do, is low and not when it is high.

The second motivational state, energized to is defined by Parker et al. (2010) as “the key direct pathway influencing goal generation and striving across a range of proactive goals” (p. 839). An important concept in this motivational state is core affect, which “refers to momentary, elementary, feelings that combine both valence and activation” (Parker et al., 2010, pp. 838-839). Positive affect enables individuals to set more challenging goals and to engage with a more problematic future. In addition, it is suggested that activated positive affect, such as feeling enthusiastic, is more important than inactivated affect, such as feeling contented, for stimulating proactive action. Because, the amount of effort put into a behavior by increasing the experience of energy is increased by a high degree of activation (Parker et al., 2010, p. 839).

(17)

 

The energized to motivational state can be measured by several concepts, for example vitality, sleep quality, and sleep quantity. Fritz, Lam, and Spreitzer (2011) refer to vitality as “having significant energetic resources and is manifested in feeling enthousiastic and alive” (p. 28). In this study vitality will be used to measure the energized to motivational state. It is assumed when negative work events occur an employee will not voice their constructive change ideas and taking charge of the situation will be less, especially when the energized to motivational state, i.e. vitality, is low. However, it is expected the opposite will happen when vitality is high. Therefore, the hypotheses are;

Hypothesis 5c: Negative work events will be negatively related to voice when an individual’s motivational state, energized to, is low and not when it is high.

Hypothesis 5d: Negative work events will be negatively related to taking charge when an individual’s motivational state, energized to, is low and not when it is high.

 

       

(18)

3. Research methods

3.1 Method

This research followed a survey strategy; which enabled the collection of quantitative data obtained by the use of questionnaires. The method to investigate the research questions is a diary study. This method gained popularity over the past decade in work and organizational research (Ohly, Sonnentag, Niessen, & Zapf, 2010). This method enables researchers to investigate “thoughts, feelings, and behaviors within the natural work context as well as characteristics of the work situation, which may fluctuate on a daily basis” (Ohly et al., 2010, p. 80). The aim of this diary study is to measure the occurrence of negative work events, the change and development of employees’ affective and motivational states, and employees’ daily proactive behavior. The benefit of a diary study is the possibility to study change and development in the employees’ affective and motivational states over a certain time span (Ohly et al., 2010; Saunders, Lewis, & Thornhill, 2009). While on the contrary, a cross-sectional study explains the occurrence of the employees’ affective and motivational states or how the different factors are related (Saunders et al., 2009), but not how they change and develop which is of importance for this study.

The diary study was conducted over the course of eight workdays. The first group to participate in the diary study started on April 7th 2014 and finished on April

16th 2014. The second group started on April 14th 2014 and finished on April 24th 2014. On day 1 a general questionnaire was completed and from day 2 until day 8 the daily questionnaires were completed. Participants were asked to fill in the surveys on three moments each day. The daily questionnaires were completed in the

(19)

 

morning between 06.00 a.m. – 10.00 a.m., at noon between 10.30 a.m. – 02.00 p.m., and in the afternoon between 02.00 p.m. – 08.00 p.m. The questionnaires were programmed in Qualtrics, which is an online data collection tool. After the data was gathered the Statistical software Package for Social Sciences (SPSS) allowed statistical data analyses. The Process tool by Hayes (2013) was used to perform the mediation and moderation analyses. This is a modeling tool for SPSS, which Hayes (2013, p.2) constructed to integrate different functions of existing and popular tools.

3.2 Participants

The population of interest to this study consisted of employees, who worked at least 32 hours a week, regardless of the sector in which these employees work. The criteria are of importance due to ensuring that the time spent at work is large enough to measure change and development in affect and motivational state after experiencing negative work events in between the daily questionnaires. The minimum amount of participants needed to be at least 20 for the data to gain statistical power in a diary study. These participants were part of my personal network (see Appendix I for the invitation letter). For this diary study I used a non-probability sampling technique, namely convenience sampling (Saunders & Lewis, 2012). With convenience sampling the researcher creates a sample based on participants who are "convenient"

sources of data (Battaglia, 2013). The choice for convenience sampling enabled me

to select participants to whom I have a personal attachment. The personal attachment is beneficial due to the high commitment needed of the participants to complete the eight-day diary study. I expect participants to feel more committed by answering the questionnaires three times a day when they have a personal relationship with me. However, convenience sampling could cause sample selection bias. In addition, in Germany three other students have also acquired a sample, which led the total

(20)

number of participants in the end to be 105.

Before the data collection phase started the participants received an instruction about what was expected from them (see Appendix II for the instruction letter). During the data collection phase an e-mail was sent 15 minutes prior to the start of the time range to remind the participants to fill in the questionnaire.

3.4 Measures

Demographic variables – The demographic variables measured in this research are age, gender, education level, years of employment, years of stay in the organization, type of occupation, leadership position, and duration of average workday. The demographic variables were only measured once in the general questionnaire, which was completed by the participants on day 1. Scholars have frequently controlled for age, gender, education level, and leadership position in past studies (Bledow et al., 2013; Den Hartog & Belschak, 2007; Fritz & Sonnentag, 2009). In this study these variables are used as control variables. The participant’s age, gender, educational level, and leadership position do not matter in the relationships shown in Figure 1. By controlling for these variables their influence or interact with the mediator, moderator, and outcome variables is eliminated.

Proactive personality – Proactive personality was measured by the shortened version of Seibert, Crant and Kraimer (1999) 10 items scale to measure proactive personality. Proactive personality was measured once in the general questionnaire. The Cronbach’s alpha equals .83 with all items included and is at its highest with all items included. Examples of items for proactive personality are: “I am constantly on the lookout for new ways to improve my life”, “Where ever I have been, I have been a powerful force for constructive change”, and “No matter what the odds, if I believe in

(21)

 

something I will make it happen”. The participants are asked to indicate for each statement on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).

In this study proactive personality functions as a control variable. I have chosen to include proactive personality as a control variable, because as mentioned in the literature review this concept of proactive behavior is a stable dispositional variable and not specifically a behavior (Bindl & Parker, 2010; Tornau & Frese, 2013). In this research I focus on proactive actions, namely voice and taking charge. Since, proactive personality, voice, and taking charge are to some extent related and interact with each other (Tornau & Frese, 2013) I want to make a clear distinction and avoid interaction with proactive personality.

Negative work events – Negative work events are assumed to provide a reason to change the status quo and show proactive behavior. Negative work events such as hindrances in goal attainment, obstacles in completing work tasks, overload; conflicts and communication problems; ambiguity, insecurity (Ohly & Schmitt, 2013) will be assessed. In this study the five items used to measure the occurrence of negative work events are selected from Ohly and Schmitt (2013) their work events checklist. In addition, the negative work events measured at noon are used to test the hypotheses. The work events checklist includes five negative work events with a Cronbach’s alpha of .76 when all items are included. When removing item 1 (which is “Did you experience time pressure, excessive demands, or did you recognize mistakes which resulted in difficulties to fulfill your work tasks?”) of the scale then the Cronbach’s alpha increases to .79. However, all items are included, since the Cronbach’s alpha of .76 is also a reliable measure and the difference is marginal. An example of a negative work event is: “Managerial and internal problems, organizational climate: Did you experience any situation that negatively affected the work climate and the

(22)

cooperation among the people in your department/your company (e.g., dismissal of a colleague, issues dealing with the supervisor, unsuccessful team meetings)?” The participants are asked to indicate for each question whether the negative event occurred or not, therefore the response scale is on a yes or no basis. In addition, the participants also get the chance to state negative work events that are not mentioned in the questionnaire.

Affect regulation – Affect regulation consists of shifts between positive and negative affective states that individuals can experience within a time interval (Bledow et al., 2011). To be able to measure affect regulation positive and negative affect is measured at multiple times during the day. In my study I will look at the affective state at noon (t2) and in the afternoon (t3). For the analyses I created a differential term

for positive and negative affect in order to measure the difference over time. To understand what high/low values of the differential terms mean for both positive and negative affect the following examples will give more insight. For positive affect, when a person states to feel high positive affect, a 5 on the scale, at noon and in the afternoon low positive affect, a 1 on the scale, the change over that time period is minus 4. Here a low value of the differential term entails that positive affect regulation was not effective from noon to afternoon, because it decreased. The opposite is the case for a positive differential value. In the case of negative affect when a person states to feel high negative affect, a 5 on the scale, at noon and in the afternoon low positive affect, a 1 on the scale, the change over that time period is also minus 4. However, a negative differential term for negative affect means that it is effectively regulated from t2 to t3 and a positive differential term of negative affect

means that it is ineffectively regulated. Therefore, I recoded the differential term for negative affect; the minus signs were changed to positive signs and vice versa.  

(23)

 

In this study the fifteen items to measure positive and negative affect are based on the scale by Kessler and Staudinger (2009). The 15 adjectives range from negative to positive affect, like exhausted, fatigued, spent, enthusiast, relax, and proud. The participants have to indicate for each adjective to what extent they strongly disagree or strongly agree on a 5-point scale.

The Cronbach’s alpha for the 6 items of positive affect measured at noon (t2) equals .88, however if item 4 is deleted the value of α increases to .90. Since, this is not such a large increase I decided to include all items. However, the items for positive affect can be divided in two categories, namely positive activating affect (α=.93) and positive deactivating affect (α=.88). Since, the Cronbach’s alpha for the entire scale lies between the subscales I decided to incorporate the value of Cronbach’s alpha for the entire positive affect scale. The Cronbach’s alpha for positive affect measured in the afternoon (t3) equals .89 here the same counts for the noon measurement, if item 4

would be deleted that the value of α increases to .90. In this case I also made the same decision to include item 4. The values for positive activating affect equals .93 and positive deactivating affect equals .88.

The Cronbach’s alpha for the 9 items of the negative affect scale measured at noon (t2) equals .89 and would not become higher if items would be deleted. The scale

of negative affect consists of three subscales with each three items, namely negative activating affect (α=.85), negative deactivating affect (α=.94), and self-conscious negative affect (α=.84, if item 9 would be deleted then α=.86). The negative affect scale measured in the afternoon (t3) has a Cronbach’s alpha of .84. The subscales’

Cronbach’s alpha for negative activating affect is .82, negative deactivating affect is .89, and self-conscious is .79. For all these scales it holds that deleting an item would not increase the Cronbach’s alpha value.

(24)

Energized to – Energized to is assessed through the subjective vitality scale by Fritz, Lam, and Spreitzer (2011), who adapted their scale from the original introduced by Ryan and Frederick (1997). Energized to was measured daily in the afternoon questionnaire. The scale consists of three vitality items (α=.98), namely “This morning I feel alive and vital”, “This morning I have energy and spirit”, and “This morning I feel awake and alert”. The items are scored on a 7-point Likert-scale with 1 being strongly disagree to 7 strongly agree.

Can to – The can do moderator variable is measured by role breadth self-efficacy. Role breadth self-efficacy (2 items; α=.97; based on Parker, 1998) captures “employees’ perceived capability of carrying out a broader and more proactive set of work tasks that extend beyond prescribed technical requirements” (Crant, 2000, p. 442). The measures used of role breadth self-efficacy in this study were measured at noon. The two statements participants have to answer are: “How confident are you when you have to set up new procedures for your work field?” and “How confident are you when you are trying to find a solution for a long term problem which you are analyzing?” For role breadth self-efficacy a 7-point scale is used to indicate the degree to which you agree ranging from 1 (strongly disagree) to 7 (strongly agree). Employees’ daily proactive behavior – Proactive behavior is “about making things happen, anticipating and preventing problems, and seizing opportunities” (Parker et al., 2010, p. 827). In this study proactive behavior is measured on a daily basis in the afternoon to see the changes and developments when negative work events occur. Proactive behavior is measured by voice and taking charge, which consist of 6 items in the afternoon questionnaire. The three items to measure voice have a Cronbach’s alpha of .93. The items for taking charge are based on the scale

(25)

 

from Fritz and Sonnentag (2009) and the Cronbach’s alpha equals .92. An example statement for voice is “Today, I developed and made recommendations concerning issues that affect my work group” and for taking charge; “Today, I tried to adopt improved procedures for doing my job”. The participants indicate for each statement on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Table 1 provides an overview of the measures used in this study and the point of measurement. For an overview of the questions asked in the questionnaires see Appendix III.

Noon (t2) Afternoon (t3)

General questionnaire - Current affective

states (How do you feel right now?) - Current level of role-breadth self-efficacy - Work events

checklist: Negative work events

experienced in the morning/the last hours at work

- Current affective states (How do you feel right now?) - Vitality - Work events checklist: Negative work events

experienced since the last questionnaire/in the afternoon - Daily proactive behavior (Voice, Taking Charge) - Demographics - Proactive Personality

Table 1 Overview of the measured variables

 

     

(26)

4. Results

In this chapter the results from the diary study are discussed. As mentioned in the method section the data from the diary study consisted of repeated observations in the morning, at noon, and in the afternoon. Before computing for each participant the aggregated mean scores of the variables over the repeated observations of 7 workdays, I recoded the values of the negative work events variable. Since 1 meant that a negative work event occurred and 2 that no negative work event took place. This was changed into 1 meaning yes and 0 meaning no, because the higher the value the more likely it occurred and not the other way around. After, computing the aggregated means I performed scale reliability analysis in order to check whether the scales are consistent in measuring the constructs (Field, 2009). Each scale’s Cronbach’s alpha was above .76 (see Table 2 for an overview). This showed that the items of the scales were able to consistently measure the construct, since a Cronbach’s alpha of .7 is a good indication of reliable scales (Field, 2009). Finally, I created the scales as the mean value of all the items on that scale.

(27)

 

Following the reliability analysis and the creation of scale means, Table 3 shows the correlations, the mean and standard deviation of the variables that were measured in the study. The correlations represent between-person correlations based on the mean scores for each participant aggregated over the repeated observations. Table 1 illustrates that negative work events is not significantly correlated to voice (r = -.07, p > .05) and taking charge (r = -.15, p > .05). When taking the relationship between the negative work events and the mediating variables, positive and negative affect at noon and in the afternoon into account the correlations with negative work events are also not significant (positive affect noon r = -.14, p > .05; negative affect noon r = .14, p > .05, positive affect afternoon r = -.13, p > .05, negative affect afternoon r = .17, p > .05). The relationship between the mediator variables and the two proactive behavior constructs, voice and taking charge, only positive affect afternoon is significantly correlated at the 0.01 level (2-tailed), however the correlation (r = .20) is not strong. In the mediating relationship negative work events is not significantly correlated to the motivational states constructs, vitality and role breadth self-efficacy (vitality r = -.10, p > .05; role breadth self-efficacy r = .03, p > .05). However, the correlation between role breadth self-efficacy and voice is .33 (p < .01), which shows a medium effect. In addition, role breadth self-efficacy is significantly correlated to taking charge (r = .20, p > .05). The correlations between vitality and the proactive behavior constructs, on the other hand, are not statistically significant (for both voice and taking charge r = .17, p > .05). In the end, I controlled for the proactive personality variable, because it influences one of the outcome variable’s constructs, namely taking charge, the moderator role breadth self-efficacy, and positive affect at noon. Proactive personality is statistically significant correlated with voice (r = .29, p > .01), role breadth self-efficacy (r = .30, p > .01), and positive

(28)

affect at noon (r = .19, p > .05); however, these correlations are equal to medium effect or less. Proactive personality explains, for example, with a correlation of .30 9% of the variance of role breadth self-efficacy. The other four control variables are demographic variables, of which gender and age significantly correlate with vitality (gender r = -.22, p > .01; age r = .21, p > .01). Leadership position, in addition, correlates significantly with voice (r = .27, p > .01), vitality (r = .20, p > .05), and role breadth self-efficacy (r = .21, p > .05). Even though, the correlations between these variables are of medium effect, I choose to control for them in the mediation and moderation analyses.

(29)
(30)

The hypotheses were tested in SPSS, using the Process macro by Hayes (2013). The Process tool enabled me to test the mediation and moderation effects between negative work events and proactive behavior at work. First, I ran the mediation analyses with the mediating variables positive and negative affect regulation (see Figure 1 for the conceptual model). These two mediating variables are differential terms of two measuring points, namely the change in affect from noon to the afternoon. The values measured at t2 are subtracted from the values measured at t3 (t3-t2), which equals the difference over that period of time. The mediation variables show the difference in positive and negative affect from one point in time to another (for a detailed explanation of the differential terms for affect see chapter 3.4). In total four mediation analyses were done and analyzed, for a summary of the results see Table 4. As previously discussed the control variables taken into account are age, gender, education, leading position, and proactive personality.

Hypotheses 1a and 1b were tested by examining the direct effect between negative work events and the two proactive behavior constructs voice and taking charge separately in a linear regression analysis. This means that the direct relationship was examined twice, namely negative work events in relation to voice and negative work events in relation to taking charge. The hypotheses stated that negative work events are negatively related to voice and taking charge. The direct relationship between negative work events and voice is not significant (p > .05), but shows a moderate negative relationship with a β-value of -.49. However, the negative work events explain 15.8% (p < .05) of variance in voice. The second, direct relationship between negative work events and taking charge is also insignificant (β = -.44, R2=.06, p > .05). Thus, hypotheses 1a and 1b are not supported, although if the

(31)

 

p-values would have been smaller than .05 the negative relationship between negative work events and the two daily proactivity constructs would have been supported.

The second set of hypotheses (H2a,b) were tested by examining two relationships, namely (1) negative work events and positive affect regulation, and (2) negative work events and negative affect regulation. The analysis for the first relationship revealed that people who experienced negative work events down-regulated their positive affect with 14.7% from noon to the afternoon. The relationship between negative work events and positive affect regulation is not significant (β =-.147, p > .05, CI [-.49, .19]). In turn, the strength between negative work events and negative affect regulation is depicted by a trivial negative β-value of -.01, which means that after the experience of negative work events negative affect was down-regulated by 1%. Unfortunately, this relationship is not significant (p > .05, CI [-.23, .25]). Thus, hypotheses 2a and 2b are not supported.

The third set of hypotheses (H3a,b,c,d) states four relationships. Hypothesis 3a states that positive affect regulation will be positively related to voice. The data shows that those who effectively regulated positive affect expressed 20% more voice, but it is not significant (β = .20, p > .05), and the confidence interval [-.20, .61] included zero. Hypothesis 3b explains the relationship between positive affect regulation and taking charge. From the data it became clear that people who effectively regulated positive affect showed .4% more taking charge behavior (β =.004), which is a marginal increase and the relationship is not significant (p > .05, CI [-.37, .38]). Hypothesis 3c states that negative affect regulation will be negatively related to voice, meaning that individuals who ineffectively regulated negative affect will not engage in the proactive behavior voice. The data shows when negative affect is regulated ineffectively expressing voice will decrease with 5%. However, the relationship is

(32)

insignificant, and the confidence interval includes zero (β = -.05, p > .05, CI [-.63, .53]). Hypothesis H3d states that negative affect regulation and taking charge are negatively related. In contrast to hypothesis 3c, when negative affect is regulated ineffectively taking charge increases with 12.1%. However, the relationship is not significant (β = .121, p > .05, CI [-.41, .65]). Thus, statistically all four hypotheses are not supported.

Hypotheses 4 states that positive and negative affect regulation will mediate the relationship between negative work events and daily proactive behavior. Since, hypothesis 3 consisted of four hypotheses hypothesis 4 has to be broken down too. The first analysis in which the mediation affect is investigated is with positive affect regulation as the mediator and voice as the outcome variable. The direct effect of negative work events on voice showed that the experience of negative work events decreased the expression of voice with 46.2%, however the results are statistically insignificant (p > .05, CI [-1.15, .23]). The indirect effect through which negative work events influence the expression of voice by positive affect regulation is relatively small compared to the direct effect and insignificant (β = -.03, CI [-.27, .03]). In the second analysis positive affect regulation as the mediator and taking charge as the outcome variable were investigated. The direct effect showed that those who experienced negative work events initiated less behavior to taking charge (β = -.44, p > .05, CI [-1.08, .20]). The indirect effect of negative work events through positive affect regulation on taking charge was marginal and not significant (β = -.0006, CI [-.08, .08]). The third analysis considered negative work events as the predictor, negative affect regulation as the mediator, and voice as the outcome variable. The direct effect between negative work events and voice showed to be that when an individual experiences negative work events they are inclined to express less

(33)

 

voice (β = -.49), but the relationship is statistically not significant (p > .05, CI [-1.18, .20]). The indirect effect through which negative work events influence voice via negative affect regulation is marginal and insignificant (β = -.0005, CI [-.08, .09]). The final analysis included negative work events as the predictor, negative affect regulation as the mediator, and taking charge as the outcome variable. The direct effect, which is negative work events influence the taking charge behavior. From the analysis it is shown that negative work events decrease the behavior of taking charge (β = -.44), however this relationship is not significant (p > .05, CI [-1.08, .20]). The indirect effect entails that negative work events influence taking charge via negative affect regulation. This effect is relatively small compared to the direct effect and insignificant (β = .001, CI [-.06, .07]). In the end, the results from the direct and indirect effects all show that there is no significant mediation effect in either of the relationships. Therefore, hypothesis 4 is not supported.

(34)

The last set of hypotheses (5a-d) was tested, by running a moderation analysis in Process. Results are presented in Table 5. Hypothesis 5a states that negative work events will be negatively related to voice when an individual’s motivational state, can do, is low and not when it is high. The can do motivational state is measured by the item role breadth self-efficacy. I examined hypothesis 5a by the incremental contribution of the interaction between negative work events and role breadth-self efficacy in voice after controlling for the main effects of negative work events and role breadth self-efficacy, and the control variables. In this relationship negative work events and role breadth self-efficacy explain 20.1% of the variance of voice (p = .005). However, the interaction between negative work events and role breadth self-efficacy is not significant (β = -.06, ∆R2 = .0004, p > .05; Model 1 in Table 5). Thus, the role breadth self-efficacy does not serve as a moderator and the proposed statement that the effect of negative work events on voice is moderated when role breadth self-efficacy is low is not correct. Therefore, hypothesis 5a is not supported.

Hypothesis 5b states that negative work events will be negatively related to voice when an individual’s motivational state, energized to, is low and not when it is high. The energized to motivational state is measured by the item vitality. I examined hypothesis 5b by the incremental contribution of the interaction between negative work events and vitality in voice after controlling for the main effects of negative work events and vitality, and the control variables. The variance explained by negative work events and vitality in voice equals to 17.7% (p = .01). Nevertheless, the interaction between negative work events and vitality is not significant (β = -.12, ∆R2 = .002, p > .05; Model 2 in Table 5). The role of vitality does not serve as a moderator and hypothesis 5b is not supported.

(35)

 

Hypothesis 5c states that negative work events will be negatively related to taking charge when an individual’s motivational state, can do, is low and not when it is high. This hypothesis was examined by the incremental contribution of the interaction between negative work events and role breadth self-efficacy on taking charge after controlling for the main effects of negative work events and role breadth efficacy, and the control variables. Negative work events and role breadth self-efficacy explain 7.7% of the variance of taking charge, however, the relationship is not significant (p > .05). Accordingly, the interaction between negative work events and role breadth self-efficacy on taking charge is statistically insignificant (β = .08, ∆R2 = .001, p > .05; Model 3 in Table 5). Thus, the role of role breadth self-efficacy does not serve as a moderator and hypothesis 5c is not supported.

The final hypothesis (H5d) states that negative work events will be negatively related to taking charge when an individual’s motivational state, energized to, is low and not when it is high. I examined hypothesis 5d by the incremental contribution of the interaction between negative work events and vitality on taking charge after controlling for the main effects of negative work events and vitality, and the control variables. Negative work events and vitality explain 8.5% of the variance, but it is not significant (p > .05). Consequently, the interaction between negative work events and vitality on taking charge is not significant (β = -.03, ∆R2 = .0002, p > .05; Model 4 in Table 5). In addition, the conditional effects of negative work events on the proactivity variables were for each hypothesis at the 10th, 25th, 50th, 75th, and 90th percentiles not significant. Thus, the role of vitality does not serve as a moderator and the proposed hypothesis that the effect of negative work events on taking charge is moderated when vitality is low is not correct. Therefore, hypothesis 5d is not supported.

(36)

The remainder of this thesis contains the discussion of the results, which is the following chapter. Subsequently, the limitations of the study are depicted. Third, the recommendations for future research are given. Final, the conclusion is stated.

(37)

 

5. Discussion

The goal of this study was twofold; first, the mediation effect of positive and negative affect regulation was investigated between negative work events and proactive behavior relationship. Second, the moderator effect of the can do and the energized to motivational states on the relationship between negative work events and proactive behavior were studied. The link between negative work events and proactive behavior has not been investigated in relation with positive and negative affect regulation, and can do and energized to motivational states (Bindl & Parker, 2010; Bledow et al., 2011; Bledow et al., 2013; Fritz & Sonnentag, 2009; Parker et al., 2010). In addition, negative affect is frequently conceptualized as the opposite of positive affect; as a result researchers did not include it in their study (Bledow et al, 2013). In this chapter the findings of the study will be discussed and related to theory and practice.

The mediated and moderated effects were both found not to be significant. In total 13 hypotheses were tested. In the analyses of hypothesis 1a and 1b, I found that when negative work events occur the participants expressed less voice and showed less taking charge behavior. However, the results of this study are not statistically supported. In previous research the opinions are divided regarding negative work events being a barrier, which distracts employees from their work. In this research the assumption was made that negative events at all times act as a barrier. Employees consider situational constraints as challenging and keeping them away from engaging in proactive behavior. However, a solution to overcome the situational constraints is challenge appraisals, because these appraisals are found to promote proactive behavior (Bindl & Parker, 2010). Another reason why this assumption was not statistically supported could be, because a distinction between short-term and

(38)

long-term effect needs to be taken into account (Bledow et al., 2011; Fritz & Sonnentag, 2009). In the daily course of action this means for practitioners that when negative work events, like problems in interaction with clients, and conflicts and communication problems with colleagues or supervisors, occur the likelihood of an employee to point out misconduct and recommend changes is small when the negative event has recently occurred. There is a possibility that when time passes and the situation is not changed that employees do start to communicate their opinions. Therefore, scholars need to explore which types of appraisals or other kinds of strategies could motivate employees to behave proactively short after negative events have occurred.

The relationship between negative work events and positive and negative affect regulation has been assessed. The analyses showed that the hypotheses were not significant, and the effects of negative work events on positive affect were small and trivial for negative affect regulation. The reason for these small changes can be a result of the nature of the two affect regulation measures, because they are a differential term between noon (t2) and the afternoon (t3). The differences between

these two measurement points are marginal. Other research methods need to be investigated to capture affect regulation. I tried to control for positive and negative affect at t2 and use the same variables measured at t3 as the mediator. The results

were also statistically insignificant. Even though, following the affective events theory it is important to assess negative work events as proximal antecedents of affect. The research conducted by Ohly and Schmitt (2013) showed that the occurrence of negative work events led to higher levels of negative affect. Key is that in their research the negative work events are clustered and the effect of each cluster was individually tested on negative affect. In my study I did not take each individual

(39)

 

cluster into account during the analysis. By investigating each negative work event cluster individually the effect of negative work events might have shown partial significance, because not all clusters have the same impact on affective states.

In turn, the effect of positive and negative affect regulation on proactive behavior has been investigated. The results were statistically not significant and the effect of positive and negative affect regulation on proactive behavior were respectively trivial and small. Even though, positive affect regulation showed signs of being positively related to voice and taking charge. Negative affect regulation appears in contrast to my assumptions to have a positive effect on taking charge and as I expected to have a negative relation with voice. The potential positive effect of negative affect regulation on taking charge is actually not surprising. Scholars have stated that negative affect draws attention to things that are not going as they are supposed to go (Bledow et al. 2011). Due to self-regulation processes people start to analyze information, which leads to gaining a realistic and clear overview of the situation and the discrepancies. Thus, the negative affect gives the person a signal that the completion of work related tasks are being disturbed. Once realized that work related tasks are being hindered from completion the person might engage in proactive behavior by taking charge of the situation.

However, how come negative affect regulation does not positively influence voice? Taking charge and voice are both conceptualized as promotive and constructive proactive behaviors to improve the status quo. The difference might be that taking charge is action-oriented and voice communication-oriented. The effect of negative affect regulation on proactive behavior leads to an active attitude and goal-directed action (Bledow et al., 2011; Den Hartog & Belschak, 2007) rather than a communication attitude. People probably prefer to take action and change the status

(40)

quo than having to discuss the misconduct with peers and/or supervisors. Future research could address this matter to gain more knowledge on the influence of negative affect regulation on taking charge and voice.

The mediation effect of positive and negative affect regulation on the relationship between negative work events and proactive behavior is not showed. The direct and indirect effects of negative work events on the two proactivity variables were statistically not significant. Moreover, the moderation effect of two motivational states (vitality and role breadth self-efficacy), on the relationship between negative work events and the proactive behavior has showed not to be significant. A possible explanation that there is no mediator and/or moderator effect could be that negative work events were considered as continuous variables. However, the negative work events were measured as categorical variables, the clusters of negative work events occurred or not. Due to using aggregates of the daily questionnaires to create one mean for each participant, the nature of the negative work events variables was changed and could gave affected the data analyses for each hypothesis. Second, with regard to the mediation analysis, as explained earlier the difference between positive (negative) affect at t2 and at t3 is so small that its

impact on the relationship between negative work events and the proactivity variables is statistically difficult to prove. Yet, a possible explanation why the moderators (vitality and role breadth self-efficacy) do not influence the negative work events – proactive behaviors link is that the moderators do not relate to the variables in this relationship. Even though, Parker et al. (2010) state that role breadth self-efficacy is related to proactive work behavior. Scholars should further investigate these relationships and use motivation literature to build a stronger foundation for the motivational states.

Referenties

GERELATEERDE DOCUMENTEN

Abstract 41 Introduction 43 Results 44 Dark field microscopy 46 Cryo-TEM of CH-Leu and CH-Abu at time interval 46 Time profile of SAXS of CH-Leu 48 Crystal structure and

The target actor receives this information from the linking and implementing actors as well as from other sources and this pushes them to pay attention to the coming grid electricity

This model hypothesises that the digital and social are related for similar (economic, cultural, social and personal) types of fields. The influence of offline exclusion

By studying the everyday mobilities of Latino gay men in New York City and Turkish and Moroccan descent gay men in Amsterdam, this paper seeks to understand how bicultural gay

This method will collect information pertaining to the answers of the second and third research questions: how do discourses of heteronormativity, patriarchy and

The forecast skill for both snowmelt floods and snow ac- cumulation generated low-streamflow events decreases from a lead time of 8 days, which indicates a decreasing skill of

Evaluation studies show that alternatives such as disco buses and cheaper public transport have a positive effect on road safety figures (see also &#34;Why was there a temproary