• No results found

The moderating influence of personality on the relationship between job demands and job stress

N/A
N/A
Protected

Academic year: 2021

Share "The moderating influence of personality on the relationship between job demands and job stress"

Copied!
54
0
0

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

Hele tekst

(1)

Master thesis Business Administration – Leadership & Management

Author: J. P.R. Kikkert (10667288) Supervisor: Mw. Dr. W. van Eerde Date: March, 12th, 2015

Concept: Final version

The moderating influence of personality on the relationship

between job demands and job stress

(2)
(3)

Statement of originality

This document is written by Jorine Kikkert who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(4)
(5)

Acknowledgements

"

This thesis represents the final step in completing my Master Degree. The process of writing this thesis was novel for me and I found it very interesting. I am most thankful to my

supervisor Wendelien van Eerde for her guidance, support, and for providing my thesis with useful comments. Her advice, suggestions and comments were very helpful.

I would also like to thank TMF Group and TMF Curacao N.V. for giving me the opportunity to gain experience within their organization.

Especially I would like to thank Evert Rakers, Marjet Rozema, and Ginette Booi for their help, perseverance and support by the data collection within TMF Group.

Jorine Kikkert Amsterdam, 2015

(6)

Abstract

Drawing on recent theory and research on stress, we hypothesized that work should be challenging but not over-demanding, and the degree to which something is perceived as demanding varies per person. This study aimed to examine to what extent personality traits (conscientiousness and neuroticism) do influence the relationship between job demands (work overload, time pressure, and billable hour pressure) and job stress. Data was collected via an online questionnaire distributed in different organizations (and countries) where they are obliged to write billable hours. The results showed that only work overload and billable hour pressure would result in a significantly higher level of job stress, where work overload is the strongest predictor of job stress. We also found that conscientiousness only significantly moderates the relation between work overload and job stress, such that it weakens the relation. And no support was found that neuroticism moderates the relations between work overload, time pressure, and billable hour pressure and job stress, such that it strengthens the relation. Our results show that although an individual might report a high level of job

demands (or hassles), the degree of neuroticism would determine more strongly whether this would lead to the experience of job stress. So the analysis of the data revealed that

neuroticism had a pronounced main effect, rather than moderating effect.

Key words: job demands, work overload, time pressure, billable hour pressure, job stress,

personality, conscientiousness, and neuroticism.

(7)

Table of Contents

Statement of originality 3

Acknowledgements 5

Abstract 6

List of Figures & Tables 8

1. Introduction 9

2. Literature review 12

2.1 Job demands 12

Work overload 13

Time pressure 13

Billable hour pressure 14

2.2 Stress 16 2.3 Personality 18 Conscientiousness 20 Neuroticism 21 3. Method 23 3.1 Sample 23 3.2 Measures 25 Job demands 25 Job stress 26

Conscientiousness & neuroticism 27

4. Results 28 4.1 Descriptive statistics 28 4.2 Correlations 30 4.3 Hypothesis testing 30 5. Discussion 36 5.1 Main findings 36

5.2 Limitations and strengths 38

5.3 Implications for theory 39

5.4 Implications for practice 40

6. Conclusion 42

References 43

Appendix 49

Annex A 49

(8)

List of Figures & Tables

"

Figure 1: Conceptual model Figure 2: Sample age distribution

Figure 3: Sample organizational tenure distribution Figure 4: Sample job tenure distribution

Figure 5: The 2-way standardized interaction effect between conscientiousness and

workload, in relation to job stress.

Table 1: Means, SDs, correlations and reliability coefficients

Table 2: Results of moderated hierarchical regression analysis with job stress as dependent

variable and conscientiousness as moderating variable.

Table 3: Results of moderated hierarchical regression analysis with job stress as dependent

(9)

1. Introduction

From the twenty-first century, the landscape of the work environment is changing very rapidly (Quillian-Wolever & Wolever, 2003). First, with the advent of technology, physical boundaries between work and home are removed. The concept of working “24/7” and

“always being connected” is nowadays very common. Therefore significant shifts in business have taken place, which greatly affected approaches of work, perceptions and expectations of what defines a ‘good’ employee (Quillian-Wolever & Wolever, 2003). In addition, the

research shifted to focus more on employee performance, attitudes and behaviours, rather than focusing on the overall success of the whole organization (Kehoe & Wright, 2010). Many businesses and organizations are trying to create a healthy workplace and want to cut down employee health costs (Grant, Christianson, & Price, 2007). Therefore organizations are trying to focus increasingly on the well being and health of their employees. The person-environment interaction is identified as very important by studying job stress (Quillian-Wolever & (Quillian-Wolever, 2003). And despite many ambiguities and conflicting findings, it does seem fair to say that there is sufficient evidence to conclude that stress at work can have an impact on people’s health and well being (Semmer, 2003). All working people have to deal with different job demands during their job. Due to the fast changing organizational

environments the demands in organizations are high. Theory and research on stress imply that work should be challenging but not over-demanding in terms of time, speed or environmental conditions (Semmer, 2003). The degree to which something is perceived as demanding varies per person. Especially in health psychology there is increasing interest in personality traits that may act as resources in the stress process (Vollrath, 2001). Each person has a normal equilibrium level of well being which can be predicted by personality characteristics. Therefore it is important that organizations have employees that can cope with these

demanding environments. So it is interesting to wonder why one person can cope better with such an environment than the other.

Lawyers, for example, must cope with environmental stressors including financial concerns, workload and competition (Fortney, 2000). Many law firms and other financial service organizations have the emphasis on the job demand ‘maximizing billable hours’. But these high billable hour expectations can have a negative effect on employees. Some person feel pressure to be competitive about compensation, and others feel pressure in the sense that the firm is only interested in the amount of fees he or she generates for the organization

(10)

(Parker & Ruschena, 2012). But why are some employees experiencing these negative effects of billable hour pressure more or faster than other employees? The appraisal of daily hassles, such as writing billable hours, is of significance to the understanding of how personality interacts with stress (Gartland, O’Connor, Lawton, & Ferguson, 2014). This is why we arrived at the aim of this study, because we want to enhance knowledge and increased understanding on the relationship between job demands, personality, and job stress.

This study is important for several reasons. First, the results bring new information on the three main topics job demands, personality traits, and job stress. In this research the focus is on three job demands, namely; work overload, time pressure and billable hour pressure. More studies on the three main topics individually have been called for in previous literature

(Gyoerkoes, Becker, Massoudi, de Bruin, & Rossier, 2012). Especially the existing literature around billable hour pressure is very limited. And all the existing literature around this topic is about lawyers, while there are more professions that make use of writing billable hours. Second, there is not much known about the integration and cohesion of the research topics. Much is known about personality traits and job stress, but the knowledge is scattered and not integrated with billable hour pressure. Studying the Five-Factor Model in relation to job stress and billable hour pressure enables to integrate earlier findings on these topics and enables to make a framework for future research. However, little is known about the association between the Five- Factor model and job stress defined by the demand-control model (Hintsanen, Tornroos, Hintsa, Jokela, Pulkki-Raback, & Hutri-Kahonen, 2013). There are studies around billable hour pressure that mention the causal relationship with stress (Fortney, 2000), but to our knowledge, studying how personality influences the relationship between billable hour pressure and job stress is never done before. Therefore we arrived at the third reason: the field of human resource management can use the outcomes of this study in selection, recruitment, and training to create the best person-environment fit. Fourth, the work population in general struggles with the growing demands to balance work and personal life; therefore this study can provide guidance in both analysing problems and formulating workable solutions for organizations. In Japan there is a phenomenon that is called ‘karoshi’, which means sudden death from overwork. The sped-up life styles of today can result in stress, which are a major cause of early death and therefore an important factor in health development (Nishiyama & Johnson, 1997). In several advanced market economies, the number of stress-related workers’ compensation claims and mental health claims has increased enormously (Shirom, 2003). As mentioned before, these employees experience heightened job demands for excessive work

(11)

hours, fast change in technologies, and the blurring line between work- and home life. Burnout is likely to represent a pressing social problem in the future, which ensures a major public health problem and a cause for concern for policy makers (Shirom, 2003). Also according to Karasek (1979) an integrated and good understanding of human stress is especially useful to understand future economic and political development. All the before mentioned knowledge gaps lead to the following main research question:

To what extent do personality traits (conscientiousness and neuroticism) influence the relationship between job demands (work overload, time pressure, and billable hour pressure)

and job stress?

This study is structured as follows. Firstly, this study will start with a short introduction of the literature that is important in order to understand the three different job demands, job stress, and the two moderating personality traits. Secondly, chapter 3 describes the methods used for data collection and analysing. Next to that, in chapter 4 you will find an overview of the results of the data analyses and hypotheses. Chapter 5 describes and discusses more briefly the results and main findings, gives an overview of the practical and theoretical implications, describes the limitations of the study, and gives suggestions for future researches. Finally, this study ends with the main conclusions and the answer to the research question.

(12)

2. Literature review

2.1 Job demands

From the 1950s and 1960s, fewer working hours and more leisure were expected to bring social, health and psychological benefits. But many of these expectations were disturbed when the scope of the intellectual debate shifted from leisure to burnout, work overload, time famine and stress (Haworth & Veal, 2004). Nowadays, working in an organization is often very demanding. These job demands are the organizational, physical, or social aspects of a job that require sustained effort and are therefore associated with certain costs (Kinnunen, Felft, Siltaloppi, & Sonnentag, 2011; Mackay & Perrewé, 2014). Job demands are not necessarily negative, but (chronic) job demands could relate to strain (Bakker & Demerouti, 2007). Organizational demands may reflect both challenging and threatening aspects; it is the appraisal of the demands that really matters (Mackay & Perrewé, 2014). Individuals in all types of work situations have many strategies in order to be able to respond to changing demands. Stable individual characteristics are argued to affect the primary cognitive appraisal of an organizational demand (Mackey & Perrewé, 2014). However, the more recently

personality-focused stress research has begun to focus on daily hassles (O’Connor et al., 2009). Where hassles are defined as “events, thoughts or situations which, when they occur produce negative feelings such as annoyance, irritation, worry or frustration, and/or make you aware that your goals and plans will be more difficult or impossible to achieve” (O’Connor et al., 2008, p.20). Therefore we can suggest that probably no person leads a life without any hassles. The impact of hassles on physical and mental health must depend on factors such as chronically high frequency of hassles and the heightening of hassles during a period (Kanner, Coyne, Schaefer, & Lazarus, 1981). From this standpoint, hassles might function more as events rather than stable situations, while job demands can be seen more as stable situations. In this study we focus on job demands, because of the stability of the three demands. But the job demand ‘writing billable hours’ can also be seen as a daily hassle, because it is an event that repeated (several times) every day. Regarding billable hour pressure, we don’t know how it contributes to finishing the work by the individual. It does not seem very useful for the individual but it does seem useful for the (administration of the) organization. But again, in this study we see ‘writing billable hours’ as a job demand. Next to that the job demands ‘work overload’ and ‘time pressure’ will be discussed in this research.

(13)

Work overload

All working persons have to deal with different job demands during their job, for example (daily) workload. According to the demand-control model, work overload is a primary factor related with increased job demands (Bakker & Demerouti, 2007). When job demands and pressures in the work environment exceed the skills and abilities of the employee, this lack of fit could contribute to work overload (Spielberger, Vagg, & Wasala, 2003).Work (over)load is a subjective construct reflecting an employee’s perceived work demands (on a particular day) (Ilies, Dimotakis, & De Pater, 2010). In relation to the type of work, work overload is generally defined as the amount of time of an activity or the frequency of an activity (Groenewegen & Hutten, 1991). Work overload depends on the organizational conditions, personal characteristics and work style. In this study we make use of the subjective construct, because we asked the employees about their experienced work overload. This subjective work overload is not the same as the objective work overload measured by actual data, however these constructs are related to one another. The before mentioned theory is summarized in the following hypothesis:

H1a:There is a positive relation between work overload and job stress.

Time pressure

Time pressure is one of the difficult working conditions that are a pressure intrinsic to the job (Spielberger, Vagg, & Wasala, 2003). Perceived time pressure is a situational concept,

because it is a reflection of circumstances outside of the respondent’s control and results from inordinate demands on time. The feelings of time pressure are (in basic) related to contracted and committed time, rather than voluntary time regarding work (Haworth & Veal, 2004). Factors such as long hours of paid work, limited control over job tasks, limited control over working and non-working time, lack of choice and the extent to which their time is

fragmented by frequent switching from one activity to another are some of these

circumstances. In other words, people who work long hours, but have freely chosen their work and are interested in it, can feel less time pressure than people who work shorter hours and are not interested in their work and have little control over their work. In general, this feeling of time pressure has not necessarily its origin in an actual lack of free time, but it is rather a psychological response to our social ‘sped-up’ styles of life these days (Haworth & Veal, 2004).

(14)

objective time pressure. A subjective feeling of time pressure depends on the reflection on personal circumstances, which makes it a factor that is dependent of the person. Some

research suggests that modern societies have to deal more and more with time-pressure, where some groups face considerably higher levels of time pressures than others (Haworth & Veal, 2004). Therefore it is important to know how to reduce the magnitude of such impacts. Some possibilities lie in the context (i.e. restructuring of the workplace, greater social support), and others lie in the person him- /herself (i.e. developing and strengthening some desirable personality dispositions, and skills regarding time-management).

There also exists a relationship between the two before mentioned job demands; work overload and time pressure. The analysis regarding Haworth & Veal (2004) demonstrates that feelings of time pressure are primarily embedded in heavy loads of paid and unpaid work. And there is also evidence that time pressure as a result from work overload are in relation with higher levels of stress (Haworth & Veal, 2004). Drawing on all the before mentioned findings, we state the following hypothesis:

H1b: There is a positive relation between time pressure and job stress.

Billable hour pressure

Until the middle of the 20th century, company bills were not so comprehensive as nowadays. The bills were sent infrequently and there was not much detailed information on it. They made use of a measuring methods where some factors were important: the amount of work done, degree of task difficulty, the value for the client and the client’s ability to pay (Kaveny, 2002). Since the 1950s and 1970s hourly billing became common. According to lawyers and accountants, billable hours are hours of work, which they could charge clients (Evans, Kund, & Barle, 2004). In the 1960s, analysts, management consultants and studies suggested that those who used hourly billing in determining their fees, made more than those who used other methods (Fortney, 2000). The initial interest in hourly billing comes from the desire to

maximize the earnings and clients’ preferences to pay only for the actual time working for the client (Fortney, 2001). This was the starting point for law firms to use the billing method, bill clients, and evaluate the contributions and compensations of the attorneys. Over the last thirty years, the billable hour measure became very popular in financial service firms and also the number of billable hours has increased. Salary, bonus and growth within the firm are in direct relation with the number of hours billed (Fortney, 2005).

Managers and organizations are using different names regarding their billing

(15)

bill, and other preferring to call this a target or quota (Fortney, 2000). Either way, the employee should always meet a certain expectation. In these days, the computer is an important tool for managers to track the amount of time their employee bill and collect. But beside the different names, managers and organizations are using different ways to

communicate their billing expectations. There are organizations that do not share this information to all the employees. But there are some who conscious share this information with their employees, for example by showing a monthly sheet with (non-) worked billable hours of the employees. This way of communication could create some kind of peer pressure and it can create a competitive work environment. Within law firms, the pressure for billing time and to generate business creates a competitive environment in the hope for increased performance (Fortney, 2000).

Obviously, not the whole workday can be seen as billable; only time devoted to work for a specific client can be billed to that client. There are some non-billable matters, for example training and education, client development, personal issues or general office work (Kaveny, 2002). The time devoted to non-billable work will vary from firm to firm, but in general is one third of all work time non-billable (Fortney, 2000). A negative consequence of this pressure could ensure a decrease in time spending in training and education programs. There are more negative consequences around billable hour pressure. On one hand, the amount of time employee work may cause some negative effects, for example unhappiness or stress. During the last thirty years, the pressure to work long hours is getting much bigger (Fortney, 2000). This makes it more difficult for employees to find a healthy work-life balance. Also creates this pressure a kind of ‘time famine’ because employees may do not have enough time for themselves and their social environment (Fortney, 2000). On the other hand, the way in which they understand the time they spend working may cause some negative effects. The billable hour pressure implies a distorted and negative account of the meaning and purpose of the employees’ perception of time (Kaveny, 2002). Because of their different and commodified understanding of time they could also look differently at things such as family birthdays, holidays and volunteer work (Kaveny, 2002). This can be seen as commodified, because more and more aspects of human activity and the results thereof are expressed in a monetary value rather than the intrinsic value. And this commodification of their time perception can seep into a lot of aspects of their entire life. They may find that work is the only activity that has meaning in their life and they view every hour as potentially available for work. Their perception of time basically arises from the manner in which they are forced to understand the time they spend working. This perception can ensure that the

(16)

employee becomes faster and more stressed. Therefore, we propose the following hypothesis:

H1c:There is a positive relation between writing billable hours and job stress.

2.2 Stress

Work stress is one of the topics of interest in many different research fields, industries and at different levels of analysis. Around 1960 the interest in work stress grew. The belief that work stress is a causal factor in physical and mental disorders and organizational outcomes has gained widespread acceptance (Ganster & Schaubroeck, 1991; Haworth & Veal, 2004; Thorsteinsson, Brown, & Richards, 2014). Stress can be defined as a natural interaction between a perceived challenge, or threat to one’s goal (known as the stressor) and the coping responses available to the individual (Quillian-Wolever & Wolever, 2003). Stress is the result of a situation in which the individual perceives a challenge and judges that his or her coping resources are not sufficient enough to manage the challenge (Quillian-Wolever & Wolever, 2003).

Mackey & Perrewé (2014) have recently designed an integrative conceptualization of workplace stress. The Appraisals, Attributions, Adaptation (AAA) model combines research from different models and theories according to the complex experiences of employees regarding organizational demands (see Appendix). This results in a more comprehensive examination of the job stress process than previously available. As mentioned above the AAA model integrates different models and theories regarding stress. All these different approaches concerning job stress acknowledge the important roles of individual cognition, appraisal, and resources. One of these approaches is the well-known and widely used Job- Demands- Control (JD-C) model (Karasek, 1979). This theoretical approach is the basis for many large-scale studies of job stress and demonstrated a significant impact on job stress research (Mackey & Perrewé, 2014). The model specifies two broad constructs that can vary

independently in the work environment (Ganster & Schaubroeck, 1991). The first construct is the psychosocial characteristics of work, such as requirements for work fast and hard or not heaving enough time. The other construct is a combined measure of the task control (e.g., pace and repetitiveness) and skill use (control over the use of skills), called: job decision latitude or control (Hintsanen, Tornroos, Hintsa, Jokela, Pulkki-Raback, & Hutri-Kahonen, 2013). Work stress occurs when the psychological demands of the job are high and the worker’s decision latitude in the task is low (Karasek, 1979). Karasek and Theorell (1990) updated this model by adding ‘social support’ as an important factor in determining employee

(17)

responses to job demands. This newer theory is called the Job-Demands-Resources (JD-R) model (Mackey & Perrewé, 2014). Another theory that is used in many models of

organizational behaviour (Ganster & Schaubroeck, 1991) is the Person-Environment fit model (French, Caplan, & Harrison, 1982). The main idea is that experiencing stress (or strain) depends on the degree of fit between the individual and the job environment (Ganster & Schaubroeck, 1991). According to French et al. (1982) there are two types of fit: one fit between outcomes provided by the job and the needs, motives, or preferences of the

individual person, and one fit between the demands and requirements of the job and the skills and abilities of the worker (Ganster & Schaubroeck, 1991). The PE-fit model is one of the first theories that look at the personal aspects (f.e. individual needs, values, and goals)

regarding stress. According to this theory it is important to look at the person who is doing the job, because not every person is suitable for every job.

It is recognized that there are individual differences in stress reactivity and stress responses (Hintsanen, Tornroos, Hintsa, Jokela, Pulkki-Raback, & Hutri-Kahonen, 2013). The transactional model (Lazarus, 1993) of stress is one of the most popular approaches to

understand job stress, because it focuses on how employees subjectively interpret objective environmental conditions (Mackey & Perrewé, 2014). It relies on an environmental demand and looks at the cognitive process that an employee undergoes, when the employee comes in contact with environmental (i.e. organizational) demands. It takes into account two processes, namely: cognitive appraisal and coping. Cognitive- or primary appraisal is a cognitive

evaluation of whether the organizational demand is relevant or irrelevant to the well being of an employee (Mackey & Perrewé, 2014). This process stops if the demand is irrelevant for their health and wellbeing. But if the demand is relevant and is thus perceived as threatening, challenging, or harmful the cognitive process continues with making attributions about that demand. This threat versus challenge view is better known as the Hindrance – Challenge Occupational Stressor Model (LePine et al., 2007). The ‘hindrance stressors’ are more threatening organizational demands, which stimulate negative emotions and limit personal gain, -growth, and –development. The ‘challenge stressors’ are the opposite, because these are demands that stimulate positive emotions, stimulate problem-solving coping and promote learning performance, personal gain, -growth, and –development (LePine et al., 2007). It is important to recognize that it is the individual’s appraisal of the threat or challenge that determines the response of the employee (Mackey & Perrewé, 2014). Therefore it is argued that individual differences (i.e. personality traits) may affect individual appraisals of

(18)

Next to that, the secondary appraisal process determines if anything can be done to cope with that threaten situation. In this secondary stage, individuals evaluate their available options (i.e. personal resources and job resources) to effectively cope with the experienced stressor. This stage leads to actions that are based on felt emotions, perceived personal and job resources, personal liabilities, and job constraints (Mackey & Perrewé, 2014).

Also in other areas of organizational research (Weiss & Adler, 1984) it has been found that individual differences affect stress outcomes in many ways. Persons with different

personality traits see objective (stressful) events in a different way. As mentioned before, some theoretical approaches imply that work stress is also influenced by person-based factors. The DC-R model assumes that stress is a response to the work environment and a response on how a person experiences the degree of demands and decision latitude (Karasek, 1979). Also according to the PE- fit model, personality has frequently been specified as a ‘need’ for the characteristic in this model. Previous research (Mackey & Perrewé, 2014) has identified several personal resources, including self-efficacy (Chen, Gully, & Eden, 2001), and self- regulation (Mackey & Perrewé, 2014), which have beneficial effects for employees by their cognitive (primary) appraisals and evaluations of organizational demands (Folkman & Lazarus, 1990). In short: personal resources (f.e. stable individual characteristics) make employees able to cope with different job demands.

In summary, measures of job stress provide information about a number of factors that influence stress in the workplace (Spielberger, Vagg, & Wasala, 2003). Therefore it should be noted that the mentioned theoretical perspectives differ in their conceptualization of stress and place different emphasis on individual personality differences and on situational variables that may moderate stress relations (Shirom, 2003). This research will mainly focus on the

psychological strain, primary appraisal of hassles, and personality.

2.3 Personality

There is increasing attention to the study of individual differences in the perceptions of how stressful work is (Hintsanen et al., 2013). And as mentioned above, people react differently to stressful events according to their personalities.

Many decades, researchers were searching for the best comprehensive personality framework. Since the 1980s, the Five-Factor model has been used many times, because it is consistent across different theoretical frameworks, assessment instruments and ratings from different sources (Vollrath, 2001). The Five-Factor Model of personality is an important advancement in the field of personality psychology and early research indicates that the Big

(19)

Five personality factors are important in every aspect of the stress process (Vollrath, 2001). The model distinguishes five personality factors, namely; neuroticism, extraversion,

openness, conscientiousness and agreeableness (Weiss & Adler, 1984). Each factor consists of several personality dimensions and can be explained regarding the NEO – PIR Personality Questionnaire (Vollrath, 2001). The first factor is ‘neuroticism’ and consists of the following traits: depression, self-consciousness, anxiety, angry hostility, impulsiveness, and

vulnerability. Secondly, persons high in ‘extraversion’ are more sociable and warm. The third factor ‘openness’ can be linked with the traits of originality, independence, and intellectual curiosity. Besides, conscientiousness people are well organized and planful, hate delay, have control impulses, are reliable, and are often more capable of self-discipline than others. And the last factor ‘agreeableness’ consists of the traits sympathetic, trustful, and forgiving.

The reason for using this Five-Factor model is because it provides a framework that offers a stable and extensive basis for looking at the differences in personality, and can therefore explain how personality affects stress outcomes regarding work overload, time pressure, and billable hour pressure. Not all individuals perceive these job demands in the same way and not everybody will experience job stress as a consequence. Based on the theoretical descriptions and findings, we assume that conscientiousness and neuroticism people are more related to stress. For instance, the research of O’Connor et al. (2009) showed a differential impact depending on levels of conscientiousness as it moderated the

relationships between daily hassles and health behaviours (O'Connor D. , Conner, Jones, McMillan, & Ferguson, 2009). Research into the structure of personality factor

conscientiousness provides evidence for low order facets. The latest work of

conscientiousness in relation with health behaviour has emphasized the importance of examining facet effects. The facets ‘self-discipline’ (strongly related to self- control) and ‘order’ have been identified as important positive predictors of health behaviours (Gartland et al., 2014). Next to that it has been shown in earlier research that different facets have

different associations with stress appraisals (Gartland et al., 2012). For instance, self-discipline is associated with reporting fewer daily hassles per day (Gartland et al., 2014), which could be relevant for the daily hassle ‘writing billable hours’. And for the long-term importance, a study highlights that conscientiousness is the only higher-order personality trait that can be related to mortality risk (Jokela, Batty, Nyberg, Virtanen, Nabi, Singh-Manoux, & Kivimaki, 2013). Therefore we discuss these personality traits (and facets) more

(20)

Conscientiousness

According to Schmidt and Hunter (1992) conscientiousness may be the most important trait that is seen as motivation variable in personnel psychology (Barrick et al., 1993). Of interest for this research is the idea that conscientiousness is an important moderator between the relationship of a high appraisal of job demands (or hassles) and job stress. To better

understand the mechanisms through which conscientiousness might affect this relationship, it is helpful to better explain what conscientiousness is. Conscientiousness is a personality trait reflecting qualities, which can be summarized in three dimensions: achievement orientation (being reliable and hardworking), dependability (being responsible, self-disciplined and careful) and orderliness (being planful and organized) (Barrick & Mount, 1991).

As mentioned, this suggests an important association between conscientiousness and the job demands and hassles (work overload, time pressure and billable hour pressure). A recent study by Gartland et al. (2012) was the first who showed that the appraisals of daily hassles are influenced by conscientiousness. According to a daily diary design study, high conscientiousness was found to be associated with more adaptive health behaviours in response to daily hassles (Gartland et al., 2014). Moreover, (aspects of) conscientiousness may influence on health by modifying the effects of daily stressors, such that people high in conscientiousness respond to stress by engaging in more healthy behaviour (Gartland et al., 2014). In addition, Gartland et al. (2014) claim a direct effect of higher levels of facet self-discipline that is associated with the experience of fewer overall daily hassles. Therefore the aim of this study is to test the extent to which conscientiousness (order, self-discipline and these facets combined) moderates the effect of the relation between the job demand (or hassle) and job stress. This leads to the following hypotheses:

H2: Conscientiousness moderates the relation between work overload and job stress, such that it weakens the relation.

H3: Conscientiousness moderates the relation between time pressure and job stress, such that it weakens the relation.

H4: Conscientiousness moderates the relation between billable hour pressure and job stress, such that it weakens the relation

(21)

Neuroticism

Looking at the primary dimensions of personality, neuroticism is the most important determinant of psychological distress (Bolger & Schilling, 1993). Of great interest for this research is the idea that neuroticism is also an important moderator between the relationship of a high appraisal of job demands (or hassles) and job stress. To better understand the mechanisms through which neuroticism might affect this relationship, it is helpful to better explain what neuroticism is. People scoring high on neuroticism are likely to experience more problems such as negative moods due to negative events of life (Judge, Higgins, Thoresen, & Barrick, 1999). The neurotic person is anxious, worrying, and emotional and reacts very strong to all types of stimuli (Eyesenck & Eysenck, 1985). Neurotic individuals respond to stressful stimuli with strain, but the same stimuli do not produce strain in low neurotic individuals (Hills & Norvell, 1991). Besides they are generally more dissatisfied than emotionally stable persons.

This suggests an important association between neuroticism and the job demands or hassles (work overload, time pressure and billable hour pressure). The hassles are of

significance to the understanding of how personality interacts with stress. Therefore the aim of this research is to test the extent to which neuroticism moderates the effect of the relation between the job demand (or hassle) and job stress. Based on the literature overview, we predict the following in the present study:

H5: Neuroticism moderates the relation between work overload and job stress, such that it strengthens the relation.

H6: Neuroticism moderates the relation between time pressure and job stress, such that it strengthens the relation.

H7: Neuroticism moderates the relation between billable hour pressure and job stress, such that it strengthens the relation.

(22)

The hypotheses 1 to 7 are shown in the following conceptual model (Figure 1).

(23)

3. Method

The present chapter gives insights in the characteristics of the sample, the methods to collect and analyse the data, and descriptions of the variables.

3.1 Sample

The research design of this study consists of a survey (quantitative research). The sample of this research consists of employees at different offices (in different countries) where

employees are obliged to write billable hours. To reach these people, the online survey is mainly distributed within an international trust office, called TMF Group. Because of too few respondents, the online survey is also distributed through LinkedIn, Facebook, and at

workplaces’ of acquaintances with the condition that they are obliged to write billable hours in their job.

In total, 181 employees filled in the online survey. From the 181 employees that started the online survey, 117 employees (65%) finished the survey by answering all the questions. The other 64 surveys were removed using list wise deletion, because of not

answering one or more questions. They dropped out before the test ends, and therefore one or more measurements are missing.

Of the 117 respondents, 63% is female and 37% is male. The mean age of the respondents is 28.92 years (SD-age = 8.49, ranging from 20 - 64 years). As you can see in Figure 2 the distribution of age within the sample shows that the majority of respondents is aged between 25 and 29 years old. The majority of respondents are highly (Master degree) educated. Just 6% only completed high school or equivalent, 11% only some college, 29% have a Bachelor degree, 50% a Master degree, and 2% have a Doctoral degree. Furthermore, it also turns out that 71 respondents works at an office located in the Netherlands, 19

respondents in Curacao, 20 in British Virgin Islands, 2 in Miami, and 5 respondents are located in a different country. Figures 3 and 4 show very clear that the majority of the respondents are working 0-3 years for their company and in their job. However, the mean organizational tenure is 7.36 years (SD-company tenure = 8.54, ranging from 0 – 37 years), and the mean job tenure is 5.20 years (SD-job tenure = 6.01, ranging from 0 – 36 years). Summarizing, we can say that the sample consists mainly of young female employees with therefore a short organizational- and job tenure, who are highly educated and working in an office located in the Netherlands.

(24)

Figure 2 - Sample age distribution

(25)

Figure 4 - Sample job tenure distribution

3.2 Measures

In this section the measures used in this research will be explained. This research contained three independent variables (work overload, time pressure and billable hour pressure), one dependent variable (job stress), and two moderating variables (conscientiousness and

neuroticism). Some widely used, validated, and reliable instruments were used to collect data in this study. The internal consistency of each scale was also tested by using Cronbach’s alpha coefficient with a cut off value of .7 or higher (Pallant, 2010).

Job demands

To measure the three job demands (work overload, time pressures, and billable hour pressure) we used a combination of methods. See the Appendix for all items.

The perceived work overload scale was measured by a five items proposed by Higgins, Duxbury, & Lyons (2010). Example items are ‘I feel I have more to do than I can comfortably handle’ and ‘I feel I don’t have enough time for myself’. Respondents were asked to rate on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). A higher score refers to a higher level of perceived work overload. The Cronbach’s

(26)

coefficient for measuring perceived work overload regarding this scale is .76, which means a good internal consistency.

To measure time pressure five items were used proposed by Teng, Shy, Chiou, Fan, & Lam (2010). Example items are ‘I feel high time pressure at work’ and ‘ I always feel in a hurry during work hours’. The items were scored on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). A higher score indicates a higher level of perceived time pressure. The Cronbach’s alpha coefficient for measuring time pressure is .86, which means a good internal consistency.

For the questions regarding the billable hour pressure we used the Daily Stressor Appraisal Scale (DSAS) with 8 items (Schneider, 2008). The DSAS scale yields three scores; primary appraisal (5 items), secondary appraisal (3 items) and the appraisal ratio. Originally consists this DSAS scale of 10 items, but during a piloting test of the development of the scale the researcher noted that two of the primary appraisal items did not perform in the item

analysis. Therefore they removed these items, so in consultation with the researcher we used the most recent version with 8 items that can be used looking back at daily stressors. For the rest of this present research, it is important to mention that we only make use of the primary appraisal scores, since we only measured the daily hassle of writing billable hours at one moment. Example items for primary appraisal are ‘How demanding do you think writing billable hours is?’ and ‘To what extent do you think you would need to exert yourself to deal with writing billable hours?’. An example item for secondary appraisal is ‘How well can you cope with writing billable hours?’. Respondents were asked to rate on a 7-point Likert scale ranging from 1 (not at all) to 7 (to a very large extent). A higher score indicates a higher level of perceived pressure due to writing billable hours. The Cronbach’s coefficient for measuring (primary appraisal) billable hour pressure regarding this scale is .88, which means a good internal consistency.

Job stress

The dependent variable, job stress was measured with the Perceived Stress Scale (Cohen, 1986) that evaluates the level of perceived stress during the last month. The reason for choosing this measure is because when measuring job stressors, more attention should be given to such facets as frequency and intensity (Spielberger, Vagg, & Wasala, 2003). Knowing how often the stressor was experienced makes it easier to measure the perceived severity of the stressor. Example items are ‘In the last month, how often have you felt nervous and stressed?’ and ‘In the last month, how often have you found that you could not cope with

(27)

all the things that you had to do?’. The PSS consists of 14 items, whereby four items are coded reversed. Respondents were asked to answer on a 5-point Likert scale ranging from 1 (never) to 5 (very often). A higher score indicates a higher level of perceived job stress. The Cronbach’s alpha coefficient for measuring perceived job stress is .81, which means a good internal consistency.

Conscientiousness & neuroticism

Personality traits were measured by questions according to the Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness Five-Factor Inventory (NEO-FFI)

questionnaire (Hintsanen, Tornroos, Hintsa, Jokela, Pulkki-Raback, & Hutri-Kahonen, 2013). The NEO-FFI questionnaire assesses the five personality traits: neuroticism (a greater

tendency towards feelings of worry), extraversion (a tendency towards sociability), openness (being open to new experiences), agreeableness (a tendency to be pleasant), and

conscientiousness (being careful and reliable). Each trait is assessed by 10 questions. The content validity of the NEO-FFI has been extensively tested and the internal consistency is very high. Respondents rated each item on a 5-point scale, ranging from 1 (very inaccurate) to 5 (very accurate).

In this present study two personality traits are used as moderators. First of all, we investigate to what extent conscientiousness acts as a moderator in the research model. As also mentioned in the theoretical part, we only make use of two facets of conscientiousness. So we used the 10-items of ‘order’, whereby five items are coded reversed. Example items are ‘I like order’ and ‘I want everything to be just right’. Next to that, the 10- items of ‘self-discipline’ are used, whereby also five items are coded reversed. Example items of ‘self discipline’ are ‘I am always prepared’ and ‘I need a push to get started’. The Cronbach’s alpha for order is .85, for self-discipline .86, and for the two facets combined .88. So, all the scales have a very good reliability. Secondly, we investigate to what extent neuroticism acts as a moderator in the research model. For neuroticism we made use of the total domain, consisting of 20 items, whereby ten items are coded reversed. Example items are ‘I often feel blue’ and ‘I worry about things’. The Cronbach’s alpha for neuroticism is .89, which means a good internal consistency.

(28)

4. Results

In this section the results of the data analysis will be reported. For the analysis of the data the program Statistical Package for Social Sciences (SPSS) was used. Descriptive statistics was used to measure some preliminary steps.

4.1 Descriptive statistics

Missing values – The survey consist of a structure where the first questions are about

demographics, the next questions about the moderators, than about the independent variables and the last questions are about the dependent variable. Therefore we decided to filter all the incomplete surveys, and count them as a missing. Incomplete surveys are surveys where participants drop out before the test ends, and therefore one or more questions are not answered. After filtering there were 117 valid surveys and 64 missing surveys. This means that the missing rate is 35% of total respondents (181 respondents).

Dummy variables – For the level of education a dummy variable was created, by using the values 0 or 1. After creating the dummy variable, the correlation between all variables was checked. These correlations are below 0.7, which indicated that there is no Multicollinearity within this research model. Therefore we used the variable ‘level of education’ as the control variable in our regression analyses.

Normality – After having recoded the reversed items, total scale means have been computed and Skewness and Kurtosis were calculated. For all variables, we observed that the values were close to the values expected in a normal distribution.

Outliers – To check for outliers we observed the boxplots. There were only outliers found for the variables workload (one outlier), time pressure (two outliers) and job stress (three

outliers). To find out to what extent those outliers are likely to be a problem, Pallant (2010) suggested comparing the mean and the 5% Trimmed Mean. For this study the two different means are very similar, namely: workload M 13,38 and TM 13,37; time pressure M 14,79 and TM 14,76; job stress M 24.32 and TM 24,33. Next to that the maximal Cook’s distances were smaller than 1, which suggests no problem (Pallant, 2010). Therefore we decided to not eliminate those outliers.

Reliability – To check the reliability of the items the Cronbach’s Alpha has to entail a value of 0.7 or higher. All the items used in this survey have a good reliability (Table 1). The scale means, standard deviations, inter-correlations and reliabilities of the variables are shown in Table 1.

(29)

Table 1. Means, SDs, correlations and reliability coefficients.

NOTE: n = 117. ** Correlation is significant at the 0.01 level (2-tailed), * correlation is significant at the 0.05 level (2-tailed), Reliabilities (Alpha’s) are found in the brackets.

Value labels of categorical variables: Gender: 1 = ‘Male’, 2 = ‘Female’. Level of education: 1. ‘Grammar school’, 2. High school or equivalent’, 3. ‘Some college’, 4. ‘Bachelor’s degree’, 5. ‘Master’s degree’, 6. ‘Doctoral degree’, 7. ‘Other’. Office: 1 = ‘Netherlands’, 2 = ‘Curacao’, 3 = ‘British Virgin Islands’, 4 = ‘Miami’, 5 = ‘Other’. Organizational- and job tenure are in years.

Variable Mean SD 1. 2. 3. 4. 5. 6. 7. a. b. 8. 9. 10. 11. 12. 1. Gender 1,63 ,48 - 2. Age 28,92 8,49 ,24** - 3. Level education 4,36 ,98 -,03 -,18 - 4. Office 1,73 1,08 ,20* ,27** -,27** - 5. Organizational tenure 7,36 8,54 ,13 ,75** -,33** ,21* - 6. Job tenure 5,20 6,01 ,22* ,59** -,28** ,10 ,66** - 7. Conscientiousness Total 3,58 0,55 ,25** ,20* -,20* ,35** ,29** ,15 [0.88] 7 a. C - Order 3,62 0,67 ,16 ,10 -,23* ,33** ,24* ,12 ,87** [0.85] 7 b. C - Self Discipline 3,54 0,62 ,28** ,25** -,11 ,27** ,26** ,13 ,84** ,46** [0.86] 8. Neuroticism 2,20 0,49 ,24** -,06 -,02 -,13 -,06 -,02 -,22* -,10 -,28** [0.89] 9. Work overload 2,68 0,68 ,06 -,06 ,02 ,00 -,09 -,12 -,14 -,14 -,10 ,34** [0.76] 10. Time pressure 2,96 0,77 ,03 -,05 ,17 -,13 -,16 ,22* -,10 -,17 ,01 ,15 ,64** [0.86]

11. Billable hour pressure 1,63 0,60 ,01 ,00 -,31** ,00 -,09 -,15 ,00 ,06 -,07 ,31** ,39** ,40** [0.88]

(30)

4.2 Correlations

Table 1 presents the correlations between the variables of the research model. These values (mean and standard deviation) are the transformed values, converted to values ranging from 0 to 10. As expected the correlation analysis showed that the dependent variable job stress correlated significantly with the independent variables workload (r = .47, p < 0.01), time pressure (r = .27, p < 0.01) and with billable hour pressure (r = .45, p < 0.01). With regard to the demographic variables and the independent variables, it is notable that the variable job tenure correlated only with time pressure (r = .22, p < 0.05), and the level of education correlated with billable hour pressure (r = -.31, p < 0.01) and job stress (r = -.22, p < 0.05). This indicates that higher educated employees experiencing a lower level of job stress and pressure according to writing billable hours. And with regard to the moderating variables, it is notable that neuroticism correlated with gender (r = .24, p < 0.01), which indicates that

woman are experiencing a higher level of neuroticism than men. Conscientiousness correlated with all demographic variables, except job tenure.

Based on these statistics, it seems that neuroticism is more related to the variables of the research model than conscientiousness. Conscientiousness (order, self discipline and these facets combined) did not correlate with any (in)dependent variable. Neuroticism correlates with workload (r = .34, p < 0 .01) , billable hour pressure (r = .31, p < 0.01) and also with the dependent variable job stress (r = .66, p < 0.01).

4.3 Hypothesis testing

To test the hypotheses, we used hierarchical multiple regression analyses. The first set of regressions is about testing the hypotheses (H1a, H1b, H1c) whether there is a relation between the independent variables and job stress. The second set of regressions is about testing the hypotheses (H2, H3, H4) whether conscientiousness (order, self- discipline and these combined) is a moderator in the relationship between the independent- and dependent variables (see Table 2). And the third set of regressions (H5, H6, H7) has been performed with neuroticism as a moderator (see Table 3).

As can be seen in the correlation matrix, only the level of education correlated with the dependent variable job stress. Therefore the level of education is used as a control variable in both regression analyses. The regression analysis indicated that this control variable explained 4,9% of the variance in job stress (!!= .049). Adding this control variable increased the

(31)

The results of the regression analysis indicate relationships between the independent variables with the dependent variable and the effects of the moderating variables. Hypotheses 1a, 1b, and 1c are about the relation between job stress and the independent variables.

Hypothesis 1a predicted that a high level of workload would result in a higher level of job stress. Workload was significantly positively related to job stress (! = ! .411, ! < 0.01). Therefore, H1a is supported. Hypothesis 1b predicted that a high level of time pressure would result in a higher level of job stress. Time pressure was not significantly related to job stress (! = ! −.083). Therefore, H1b is rejected. Hypothesis 1c predicted that a high level of billable hour pressure would result in a higher level of job stress. Billable hour pressure was

significantly positively related to job stress (! = ! .284, ! < 0.01). Therefore, H1c is supported. The three independent variables explained together 27,4% of the variance in job stress (!!= .274, p < 0.01). As can be concluded from the findings concerning the three job

demands, the largest stressor is work overload (! = ! .411, ! < 0.01). In conclusion: the regression analysis showed that there were significant effects that work overload and billable hour pressure will result in a higher level of job stress.

For the moderating variables we also performed hierarchical regression analyses, and hypotheses 2 – 7 relate to this. First we took conscientiousness as a moderator. Hypothesis 2 states that conscientiousness moderates the relation between work overload and job stress, such that it weakens the relation. Hypothesis 3 predicts that conscientiousness moderates the relation between time pressure and job stress, such that it weakens the relation. And

Hypothesis 4 proposes that conscientiousness moderates the relation between billable hour pressure and job stress, such that it weakens the relation. We measured only two facets of conscientiousness, namely; order and self discipline. Therefore we performed three different hierarchical regression analyses for the two facets individually, and one for the facets

combined (‘total’). When we first look at ‘order’ we see that this moderating variable

explained 2,7% of the variance in job stress (!!= .027, ns). Also ‘self discipline’ explained a

little variance, 2%, of the variance in job stress (!!= .020, ns). When we combine the facets

order and self-discipline, we see that this moderating ‘total’ variable explained 3% of the variance in job stress (!!= .030, ns). No support was found for H2; conscientiousness did not

moderate the relationship between workload and job stress, such that it weakens the relation (! = ! −.135, !"). Also no support was found for H3; conscientiousness did not moderate the relation between time pressure and job stress, such that it weakens the relation (! = ! −.052, !"). And H4 is also not supported; conscientiousness did not moderate the relation between

(32)

billable hour pressure and job stress, such that it weakens the relation (! = ! −.006, !"). In conclusion: the regression analysis showed that there were no significant moderating effects of conscientiousness (neither order, self discipline and total) on the relationship between workload, time pressure, and billable hour pressure and job stress.

Secondly we took neuroticism as a moderator. Hypothesis 5 states that neuroticism moderates the relation between workload and job stress, such that it strengthens the relation. Hypothesis 6 predicts that neuroticism moderates the relation between time pressure and job stress, such that it strengthens the relation. And Hypothesis 7 proposes that neuroticism moderates the relation between billable hour pressure and job stress, such that it strengthens the relation. As mentioned before we used the whole neuroticism factor, so we performed one hierarchical regression analysis (see Table 3). The moderating variable explained 24,5% of the variance in job stress (!!= .245, p < 0.01). However, no support was found for H5; neuroticism did not

moderate the relation between workload and job stress, such that it strengthens the relation (! = ! .019, !"). Also no support was found for H6; neuroticism did not moderate the relation between time pressure and job stress, such that it strengthens the relation (! = ! .017, !"). And no support was found for H7; neuroticism did not moderate the relation between billable hour pressure and job stress, such that it strengthens the relation (! = ! .046, !").

However, neuroticism explains quite a lot of the variance in job stress. In Table 3 it is clear to see that a high level of neuroticism will result in a higher level of job stress (! = ! .535, ! < 0.01). In conclusion: the regression analysis showed that there were no significant

moderating effects of neuroticism on the relationship between workload, time pressure, and billable hour pressure and job stress. But there is a direct relationship between neuroticism and job stress.

We also performed the moderator analysis with the SPSS macro ‘PROCESS’. PROCESS offers various tools for probing 2 and 3 way interactions and used bootstrapping (Hayes, 2013). After performing these moderator analyses with PROCESS, we found one difference in comparison with the analyses with SPSS. Hypothesis 2 predicted that conscientiousness moderates the relation between workload and job stress, such that it weakens the relation. According to PROCESS, H2 is supported. It indicates that conscientiousness moderates the relation between workload and job stress, such that it

weakens the relation (! = ! −.0233, ! < 0.05). Figure 5 presents the interaction effect of this hypothesis. The figure shows that people who are low in conscientiousness experiencing more job stress when they experience higher levels of workload. While people who are high in

(33)

conscientiousness experiencing less job stress when they experience higher levels of work overload.

Figure 5 – The 2-way standardized interaction effect between conscientiousness and workload, in relation to job stress.

Summarizing, the regression analyses showed that there were significant effects that work overload and billable hour pressure will result in a higher level of job stress. But the variable time pressure has no significant effect on job stress. Regarding the moderator conscientiousness (either order, self discipline and these combined) and neuroticism, there were no significant moderating effects on the relationship between work overload, time pressure, and billable hour pressure and job stress. Only according to the regression analysis with PROCESS, there was a significant moderating effect of conscientiousness found on the relationship with work overload and job stress. In addition, neuroticism explains quite much of the variance in job stress, so neuroticism plays an important role in relationship with job stress. 1 1,5 2 2,5 3 3,5 4 4,5 5

Low Workload High Workload

Job s tr es s Low Conscientiousness High Conscientiousness

(34)

Table 2. Results of moderated hierarchical regression analysis with job stress as dependent variable and conscientiousness as moderating variable.

NOTE: n = 117. ** Correlation is significant at the 0.01 level (2-tailed), * correlation is significant at the 0.05 level (2-tailed)

Model 1 Model 2 Model 3 Model 4 Model 5

B SE B ! B SE B ! B SE B ! B SE B ! B SE B !

Level of education -1,121 0,460 -0,222 -0,624 0,441 -0,123 -0,618 0,458 -0,122 -0,650 0,475 -0,128 -0,674 0,469 -0,133

Work overload 0,626 0,158 0,411** 0,678 0,163 0,445** 0,634 0,162 0,416** 0,670 0,161 0,440**

Time Pressure -0,113 0,149 -0,083 -0,157 0,153 -1,025 -0,109 0,151 -0,081 -0,144 0,150 -0,106

Billable hour pressure 0,222 0,074 0,284** 0,224 0,074 0,286** 0,208 0,076 0,266** 0,212 0,074 0,271**

C - Order -0,041 0,065 -0,052

Work overload x CO -0,018 0,025 -0,082

Time Pressure x CO -0,014 0,021 -0,071

Billable hour pressure x CO -0,006 0,010 -0,053

C – Self Discipline -0,080 0,072 -0,096

Work overload x CS -0,034 0,028 -0,137

Time Pressure x CS -0,007 0,025 -0,032

Billable hour pressure x CS 0,005 0,011 0,042

Conscientiousness Total -0,044 0,039 -0,093

Work overload x C -0,018 0,015 -0,135

Time Pressure x C -0,006 0,013 -0,052

Billable hour pressure x C 0,000 0,006 -0,006

!R! 0,049 0,323 0,351 0,352 0,358

(35)

Table 3. Results of moderated hierarchical regression analysis with job stress as dependent variable and neuroticism as moderating variable.

NOTE: n = 117. ** Correlation is significant at the 0.01 level (2-tailed), * correlation is significant at the 0.05 level (2-tailed)

Model 1 Model 2 Model 3

B SE B β B SE B β B SE B β

Level of education -1,178 0,483 -0,222* -0,688 0,462 -0,129 -1,004 0,380 -0,189**

Work overload 0,626 0,158 0,408** 0,321 0,135 0,209*

Time Pressure -0,101 0,145 -0,075 0,042 0,121 0,031

Billable hour pressure 0,219 0,074 0,280** 0,097 0,063 0,124

Neuroticism 0,287 0,038 0,535**

Work overload x N 0,003 0,014 0,019

Time Pressure x N 0,002 0,012 0,017

Billable hour pressure x N 0,004 0,006 0,046

!R! 0,049 0,324 0,568

(36)

5. Discussion

In this chapter, the main findings will be discussed and the limitations and strengths of the study will be described. Also the theoretical- and practical implications and suggestions for future research will be examined.

5.1 Main findings

With regard to the previous job stress literature, this study aimed to examine the effects of personality on the relationship between job demands and job stress. And the main

contribution of this study was to learn more about billable hour pressure.

Taking former findings in consideration, it was expected that a higher level of job demands such as work overload, time pressure, and billable hour pressure would be related to a higher level of job stress. The results showed that only work overload and billable hour pressure would result in a significantly higher level of job stress, where work overload is the strongest predictor of job stress. So the results make clear that when an employee

experiences a higher work overload during their job, the extent of experiencing job stress will increase. This also applies to billable hour pressure. Our finding about work overload is in line with Karasek’s demand – control model, which states that work overload is a primary factor related with increased job demands, and (chronic) job demands could relate to strain (Bakker & Demerouti, 2007). And our finding about the job demand (or daily hassle) of writing billable hours is in line with the findings of the research from Fortney (2000). They argue that the competitiveness and perception of time due to writing billable hours can be an explanation for the employee feeling stressed.

First of all, we were interested in the moderating effect of the personality trait

conscientiousness (order, self-discipline and these combined) in the relationship between job demands and job stress. Because conscientiousness is the most motivational trait (Barrick et al., 1993), we expected that this trait moderate the relation between the job demands and job stress, such that it weakens this relation. First, the results showed that conscientiousness (order, self- discipline and these combined) explained only a little (2-3%) of the variance in job stress, so there is no strong linear relationship with job stress. Contrary to what we expected, we could say that in this study conscientiousness did not significantly moderate the relations between work overload, time pressure, and billable hour pressure and job stress, such that it weakens the relation. This finding is inconsistent with the study of Gartland et al.

(37)

(2014) which provides evidence that (two facets of) conscientiousness can moderate the relationship between hassle appraisal and emotional well-being such as job stress. Next to that, the recent study of Gartland et al. (2012) showed that the appraisals of daily hassles are influenced by conscientiousness. But after performing these moderator analyses with

PROCESS, we found one difference in comparison with the analyses with SPSS. PROCESS gives a different outcome from the regression in comparison with SPSS because PROCESS offers various tools for probing 2 and 3 way interactions and used bootstrapping (Hayes, 2013). According to the results from PROCESS, we can state that conscientiousness moderates the relation between work overload and job stress, such that it weakens the

relation. According to the results from PROCESS it is clear to see that people who are low in conscientiousness experiencing more job stress when they experience higher levels of work overload. And people who are high in conscientiousness experiencing less job stress when they experience higher levels of work overload.

In addition, we were interested in the moderating effect of the personality trait neuroticism in the relationship between job demands and job stress. Because neuroticism is the most important determinant of psychological distress (Bolger & Schilling, 1993), we expected that this trait would moderate the relation between the job demands and job stress, such that it would strengthen the relation. The results showed that neuroticism explained quite a lot (25,5%) of the variance in job stress. This finding is in line with Bolger & Schilling (1993) and with Judge et al. (1999) who found that people scoring high on neuroticism are likely to experience more problems (such as job stress) due to negative events of life. However, no support was found that neuroticism moderates the relations between work overload, time pressure, and billable hour pressure and job stress, such that it strengthens the relation. Our results show that although an individual might report a high level of job demands (or hassles), the degree of neuroticism would determine more strongly whether this would lead to the experience of job stress. So the analysis of the data revealed that neuroticism had a pronounced main effect, rather than moderating effect. A possible explanation for this surprising result could be due to aspects of the methodology or the statistical techniques employed. The interpretation of a monotone interaction is difficult to detect statistically, because the effects are distributed between the main effect terms and the interaction term in a multiple regression analysis (Hills & Norvell, 1991).

(38)

5.2 Limitations and strengths

The current study has some limitations. The first limitation is related to the size and nature of the research sample. A larger sample size would increase reliability and validity as well as the generalizability of this study. It remains a quite small sample, even if the size of the sample (n = 117) meets the statistical power requirements. Besides, the sample with

respondents is relatively homogeneous. Most of the respondents are between 25 and 29 years old (see Figure 2), are highly educated, and have a short organizational- and job tenure (see Figure 3 and 4). The small differences in the demographic variables between the respondents can result in smaller statistical effects. This study does give, however, a picture of the current situation and feelings concerning work overload, time pressure, billable hour pressure, job stress and the personality traits. Further research could perform the same study design on a more diversified and larger group of respondents.

The second limitation is about the method of collecting the data. Not only the

employees of the trust office TMF Group were reached, but the online questionnaire was also distributed through social media (LinkedIn and Facebook) and via the network of

acquaintances. Probably we could assume that the distribution of the age is in relation with the way in which the questionnaire is distributed. Social media is something that is used by relatively younger people. And although the questionnaire was distributed on the condition of writing billable hours, people could self-select and thereby there is no evidence that the respondents actually write billable hours. This self-report method can lead to validity problems, so common method bias may thus have influenced some of the results. Respondents might see themselves more positive/negative than they in reality are and therefore obtain a higher score on the variables than they should. According to Spector (1994) self- reports can be seen as more critical for cross-sectional research because it allows less confidence for causal relations due to the probability that many other factors can

influence the relations. Especially personality traits are often subject to bias and distortion by the respondent (Spector, 1987). These limitations can affect the reliability of the study.

There are also some limitations concerning the questionnaire. First, the length of the questionnaire was quite long. Because of the size of the variables, there were many items. Next to that, during our analyses we only made use of the primary appraisal scores of writing billable hours, since we only measured this daily hassle at one moment. Though, the three items according to the secondary appraisals were asked in the questionnaire. For future research it is better to measure (and use) both the primary appraisal and the secondary

Referenties

GERELATEERDE DOCUMENTEN

Hagen of mijten van snoeiafval, al dan niet doorgroeid met (klim-)planten bevorderen een goed microklimaat met een grote diversiteit aan insekten en

Na 1870 verdween de term ‘tafereel’ uit de titels van niet-historische romans en na 1890 blijkt deze genre-aanduiding ook voor historische romans een zachte dood te

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

While methods that can quantify aneuploidy rates in interphase cells can be used to circumvent this bias, most of these methods cannot detect aneuploidies at the single cell

Expert Hospital 8: “A high workload can just urge you to say: “We have to do this now to finally get our workload down.” That is a route I hear. But you can also say: “No, the

The objective of this research is to determine the relationship between job insecurity, job satisfaction, affective organisational commitment and work locus of control

That is, in case of high job demands, workers with a high active coping style may be more likely to activate job resources than workers with a low active coping style (cf. For

Naast veranderingen met betrekking tot de inhoud en vormgeving van Opzij tussen 1981 en 1997, en de concurrentie die Opzij in deze periode had, zijn er ook een aantal factoren