• No results found

The world of tasks : how are jobs changing and what are the forces behind it?

N/A
N/A
Protected

Academic year: 2021

Share "The world of tasks : how are jobs changing and what are the forces behind it?"

Copied!
62
0
0

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

Hele tekst

(1)

The world of tasks

How are jobs changing and what are the forces behind it?

Name: Laura Steenman

Studentnumber: 10857974

Faculty: Economics and Business from the University of Amsterdam

Study: Master Business Administration; track Leadership and Management

Date: June 29th 2015

Supervisors: Hannah Berkers (UvA) and Luc Dorenbosch (TNO)

Abstract

The purpose of this study is to gain a better insight in what types of tasks are more open to changes and why jobs are changing. Tasks are coded into core, noncore and extra tasks by using a self-developed coding system. The first part, which is quantitative, showed that within primary school teachers mainly changes in extra tasks are made. Extra tasks are only done by some teachers and are not fundamental to the job. Furthermore, results showed that differences exist between how primary school teachers perform their job. The differences in core and noncore tasks are more seen between-teachers in comparison to within-teachers. However, in extra tasks those differences between-teachers are equal to within-teachers. In the second part, people in different jobs are interviewed to analyse why jobs are changing. It is seen that both internal and external factors influence job changes. Again most internal forces result in extra task changes, except of changes due to lack of time /efficiency or financial reasons. The results implicate that job changes are most seen in extra tasks, which means that the tasks that are

(2)

Table of contents

General introduction 4

Study 1: Task Dynamics

Theoretical framework 6

Changing jobs 7

Job analysis 8

Classifying tasks 11

Theoretical background task coding 14

Task coding 16

Hypothesis 18

Research method 21

Mooi werk tool 22

Sample 23 Procedure 24 Data analysis 25 Results 28 Descriptive statistics 28 Intraclass correlation 29

Levine’s F test for homogeinity in variances 30 Core tasks and job meaningfulness/satisfaction 31

Conclusion 33

Link to study 1 and study 2 33

Study 2: Forces behind job crafting

Theoretical framework 34

(3)

External forces 38 Research method 39 Preliminary study 39 Interviews 40 Sample 40 Procedure 41 Data analysis 41 Results 41 Internal forces 42 External forces 44 Conclusion 44 General discussion 45 Implications 47 Limitations 47 Further research 49 References 50

Appendix 1: Table with tasks 56

(4)

General introduction

An enormous amount of research is done on the concept of job analysis, because its of importance in human resource management. Job analysis is seen by most researchers as the foundation of nearly all human resources applications (Sanchez & Levine, 2012; Siddique, 2004).For example, selection tools, training and development systems, and performance evaluations (Morgeson, Spitzmuller, Garza & Campion, 2014; Sanchez & Levine, 2012). In the process of job analysis one gains an understanding of the activities, goals, and

requirements demanded when fulfilling tasks within a job (Sanchez & Levine, 2012). Job analysis can be a useful tool to acquire person-job fit, which is the match between an

employees’ knowledge, skills and abilities and the specific job requirements. It assumes that the job fits with the individual needs and that the individuals have the knowledge, skills and abilities to match the job demands (Boon, Den Hartog, Boselie & Pauuwe, 2011). A good fit can raise positive employee performance and well-being outcomes (Lin, Yu & Yi, 2014).

Even though job analysis is studied thoroughly and is used extensively in practice, it is criticised by some researchers (Morgeson & Diedorff, 2011; Sanchez & Levine, 2012). Those difficulties are related to the assumption of traditional job analysis that claims that employees, jobs and the fit between them are stable over time. Though, in the rapidly changing work environment this is not the case anymore (Grant & Parker, 2009, Grant, Fried, Parker & Frese, 2010; Singh, 2008). Furthermore, individuals do not only perform the tasks as they are intended, but also reshape their jobs to their own needs and preferences (Sanchez & Levine, 2012).This phenomen is called job crafting (Wrzesnieuwski &Dutton, 2001). Hence, it is suggested that flexibility is needed within job analysis (Singh, 2008).Job crafting and changes within the working environment are both studied extensively, however not much research is done on what specific task types are more open to changes (Singh, 2008; Petrou, Demerouti, Peeter, Schaufeli & Hetland, 2012; Tims & Bakker, 2010; Wrzesniewski &

(5)

Dutton, 2001). This gap is explained in more detail in the theoretical background. This study tries to fill this gap by answering the following question: What type of tasks are more open to job changes? This study contributes to the existing literature in three ways.

The first contribution as already mentioned is clarifying what specific task types are open to both job changes within-people over time and which are open to differences between-people in how they perform their job. It is theoretically relevant for the job analysis literature to know more about what types of tasks are more open to job crafting and changes within the working environment, because this can give more inside in how job analysis can be seen in a more flexible way. It is practically relevant for managers to know what kind of tasks are crafted/changed, since they design jobs and are responsible for the best fit between employees and jobs. As managers know more about the causes of job crafting and the way in which jobs are crafted, they get a better understanding about what employees find important within their job.

A second contribution is developing a classification system to code the tasks into different task types. In the literature tasks are coded in several ways, however those are not used in this study as is explained more thoroughly in the theoretical background. On the basis of the existing literature a new coding system is developed and used to test what types of tasks are more exposed to job changes.

A third contribution is made in the second part of this study where it is investigated why jobs are changing. Most studies focus only on a small part of possible factors. For example only internal changes such as job crafting (Demerouti, 2014; Tims & Bakker, 2010). In this study both external and internal forces behind job changes are taken into account. Employees are not guided to one answer, but they are free to talk about any factor they think plays a role in their job. Furthermore, it is analysed whether the different forces have an impact on the distinctive task types.

(6)

This research consists of two parts. The first part of the study uses a quantitative approach with secondary data of primary school teachers to examine task dynamics. It answers the questions what type of tasks changes the most within-people and whether there are differences between-people in how they execute the same job. The second part is qualitative and adds to the first study by investigating why jobs are changing and whether there are forces that influence a certain task type differently. It is investigated whether the adjustments are due to external, internal forces or both. Both parts provide a theorethical background, method section, result section and a conclusion. Finally, a overall discussion is given with the insights from both studies.

Study 1: Task Dynamics

Theoretical background

The first part of this theoretical background elaborates more on changing jobs, job analysis and explains why it is necessary to have a better understanding of changing jobs. Secondly, existing task classification systems are discussed and it is explained why a new classfication system is made. Finally, a contribution is made to the existing literature by developing a new task classficication system, it is explained how it has been created which is based on the existing literature. This new developed classification system is used for investigating task dynamics, which shows the differences in the same job between-people and how jobs have changed within-people.

Changing nature of jobs

This paragraph discusses several changes in the work environment that have influenced jobs. Firstly, technological innovations creates different work environments (Cascio, 1995). New technology changes the way things are done and those changes in turn have a personal

(7)

affect, for example where one lives and works (Burke & Ng, 2006; Cascio 1995). As technology gets better, cheaper and faster many things become easier. For example, processing and sharing of an enormous amount of information, acessing data, and

communicating. Furthermore, employees meet online or work in virtual teams which results in less travel. They are constantly checking emails, downloading reports and connecting with the office outside their working hours The use of technology increases productivity for certain business processes, because tasks become automated and the need for skilled work decreases. Moreover, technology changes the competencies required to perform jobs, because most jobs involve the use of computer technology (Burke & Ng, 2006)..

Secondly, globalization of the economy and the global competition. The growth in transportation results in a global marketplace for products and services. In addition

technological innovations lead to linking local martkets, cutting cost of moving information and breaking down all kinds of political and cultural barriers (Burke & Ng, 2006). Companies must compete for business with companies in the same industries in other countries (Cascio, 1995). Another consequence of globalization is that workers have to contend with more temporary, freelance, and contract work than in the past. However, this could have

implications for the commitment and loyalty to a company. Furthermore, companies hire and pool the most talented people in the world together, which leads to globally distributed teams that make use of conference calls in different time zones, language barriers and cultural assumptions (Burke & Ng, 2006). Moreover, organisations focus more and more on their core competences and outsourcing everthing else (Cascio, 1995).

Thirdly, there is a shift from a manufacturing to a service based economy, because the work environment becomes more information driven instead of industrial driven (Cascio, 1995). As a result of the increase in service work, employees are going beyond what is stated in their job descriptions, because they gain more roles and responsibilities. In addition, jobs in

(8)

the service sector are harder to examine objectively compared to jobs done in the manufacturing sector. This makes the use of job analysis even more complex (Sanchez, 1994).

Finally, the organisation and relationships within the organisation are changing. For example, the communication within a organization has changed and thereby the formal and hierarchical organizational structure has become less neccesary (Burke & Ng, 2006). Job analysis has focused on individual jobs. However, because of the increasing use of teams in organizations, inter-job activitities and team-based work becomes important (Singh, 2008). In addition, many organizations rotate their employees through several job posititions to enhance organizational flexibility. It asks employees to be adaptable, flexible and skilled in several domains, this is not taken into account in the traditional job analysis (Sanchez, 1994).

In sum, in the past decades much has changed in the working environment and it is interesting to take this into account when focussing on job analysis and changing tasks. These changes in the working environment are still going on and keep influencing organizations and jobs (Grant et al., 2010).

Job analysis

As mentioned in the introduction, job analysis is a preceding step in the application of many tools used in an organisation (Sanchez & Levine, 2012). Job analysis is defined by Brannick, Levine and Morgeson (2007) as:“The systematic process of discovery of the nature of a job dividing it into smaller units, where the process results in one or more written products with the goal of describing what is done in the job or what capabilities are needed to effectively perform the job” (p. 8). Job analysis is studied for a long time, but most researchers studied the reliability, validity and procedures through which job information should be gathered (Sanchez & Levine, 2012). These studies made an important contribution to the field of

(9)

human resource management by developing techniques and procedures to generate accurate and practically useful job related data (Siddique, 2004).There is an increasing recognition of the central role of job analysis in all human resource related activities, however there is little emperical research that specifically linkes job analysis to organization performance. One reason to expect the postitive relation between job analysis and performance is the central postion that job analysis occupies in human resource practices that contribute to organization performance. Siddique (2004) tested this link and found that a proactive job analysis practice is an effective human resource strategy to achieve positive organizational outcomes, such as administrative efficiency and stronger relative performance in the industry. Thus, job analysis is a tool which can affect the organization positively. Though, there are some problems in the field of job analysis that should be taken into account and those are discussed below.

One problem within the field of job analysis is that task data collected via incumbents in the same occupation often shows a lot of variance. Earlier it was thought that this is due to random or idiosyncratic error in incumbents rating (Sanchez & Levine, 2012). Though, recently reseachers are suggesting that this variance is caused by other factors. Sanchez and Levine (2012) argued that some of this variance which is often believed to be “random”, reflects systematic differences in the way some incumbents interpret and even more importantly perform their job. Moreover, Morgeson and Diedorf (2011) discussed factors that can influence incumbent rater differences in job analysis: job familiarity/job tenure, general cognitive ability, personality characteristics, work experience, performance level worker, and social and cognitive influences. Job tenure and work experience might also influence differences in how employees perform jobs over time, because they get more experienced and this possibly leads to performing the job different over time (Morgeson & Diedorff, 2011) . Another difficulty within job analysis arises from individuals who not only perform the tasks as they are intended, but also reshape their jobs to their own needs and iniatives

(10)

(Sanchez & Levine, 2012). Researchers acknowledge that employees are not passive but active recipients of the jobs they are assigned to. The earlier approach of job design that stated ‘one-size-fits-all’ is no longer sufficient (Grant & Parker, 2009). This phenomenen of changing one’s job to their own preferences is called job crafting (Wrzesniewski & Dutton, 2001). It is argued that employees craft their jobs in order to improve person-job fit and work-motivation (Tims & Bakker, 2010). Job crafting results in employees who perform the same job in different ways and a employee who performs the jobs in different ways over time.

Lievens, Sanchez, Bartram and Brown (2009) found that situational factors such as job complexity and occupational activitities influence the variance in ratings between incumbents. Occupational activities implies that some occupations are more prone to job crafting than other occupations. Moreover, they suggest that job crafting plays a role in the systematic differences found between raters. In other words, employees can be seen as active agents who modify their job to fit it to their motivation and personal goals and thereby rate the tasks differently within the same job.

Another problem acknowledged within job analysis is that the business environment is changing and therefore jobs are not as stable over time as before (Grant et al., 2010; Sanchez, 1994; Singh, 2008). Traditional job analysis assumes that jobs are statics and that job, person and the fit between them remain stable over time (Singh, 2008). Though, as already discussed before this is no longer the case. Consequences of changes in the work environment are for example teamwork and self-managing teams, which results in interrelated jobs where the completion of tasks relies also on colleagues’ efforts (Grant et al., 2010; Sanchez, 1994).  The changing working environment plays an important role in the job analysis literature, however this study does not elaborate much on this. It is hard to investigate whether the business environment influences the jobs, because a longitudal study is needed since those changes are continiously happening.

(11)

In conclusion, job analysis is a useful tool within human resource management, however some difficulties within job analysis should be considered. Those difficulties are the changing working environment, variance in incumbents ratings of job activities that are due to different interpretations and completions of the same job, and people make changes in their jobs over time (Morgeson & Diedorff, 2011; Lievens et al., 2010; Sanchez & Levine, 2012). This first study contributes to the job analysis literature by investigating whether there are really differences between-people in how they perform their job and how jobs change within-people over time. Those changes and differences may explain the problems indicated by the job analysis literature and could be taken into account in further job analysis research.

Classifying tasks

One contribution of this study is to make a classification for different types of tasks, which can be used in all kind of jobs and that can be used to study task dynamics within a job.This study categorizes on task level to get a closer look into how jobs are changing. A task is defined as “an activity that occurs in order to produce a product or outcome required on the job” (Tsacoumis &Willison, 2010, p.4).

In the job analysis literature classifications in tasks can be found, that define general categories of role requirements that are needed in a job (Morgeson & Diedorff, 2011). Examples are the classification of tasks into data, people and things (Fine, 1988; Fine & Cronshaw, 1999), and the categorization of managerial tasks into conceptual, interpersonal and technical/administrative work role requirements (Diedorff, Rubin & Morgeson, 2009). Most task classifications make a distinction between relational and task dimensions which is consistent with the role activities in the job crafting literature (Diedorff et al., 2009; Lyons, 2008). Distinguishing between relational and task related tasks might have complications, since the changing work environment (e.g. shift from a manufacturing to a services-oriented based economy) results in jobs which involve more people and relational tasks. For example,

(12)

through teamwork (Grant et al. 2010; Sanchez, 1994; Singh, 2008). In addition, this study uses data from primary school teachers, since tasks within primary school teachers mostly involve relational and internpersonal tasks those kind of categorizations are not suitable. Another approach to study task types is developed by O*NET, this is discussed in more detail below.

The Occupational Information Network (O*NET) is a comprehensive system developed by the U.S. Department of Labor and is often used in the job crafting and job analysis literature (e.g. Lievens et al., 2010; Diedorff et al., 2009). O*NET is both an

occupational classification system and a database (Peterson et al., 2001). The system provides information for 965 occupations within the U.S. economy. In order to keep the database up to date, the National Center for O*NET Development is involved in a continual data collection process aimed at identifying and maintaining the latest information on the characteristics of workers and jobs. Most of the occupational information is collected from job incumbents, including: occupational tasks, generalized work activities, knowledge, education and training, work styles, and work context areas. Occupational analysts provide the importance and level information regarding the abilities and skills associated with these occupations (Tsacoumis & Willison, 2010; Peterson et al., 2001).

In O*NET tasks per job are classified according to the below described procedure. The tasks were rated according to mean importance, relevance and frequency with which a task is performed (Tsacoumis & Willison, 2010). Importance was measured using a 5 points Likert scale. Relevance was measured in an indirect way. The incumbents rated the task as not relevant or provided a rating. In O*NET tasks are grouped into three categories: core, supplemental, and non-relevant, to help interpret the importance ratings. More specifically, statements rated on relevance or importance by 15 or more incumbents were classified into one of the three categories.

(13)

Core tasks are tasks with a relevance of more than 67% and a mean importance rating of more than 3.0. Supplementary tasks have a score of more than 67% on relevance, but a score of less than 3.0 on importance. Moreover, supplementary tasks might have a rating between 10% and 66% on relevance, regardless of the mean importance rating. Non-relevant tasks have a score of less than 10 % on relevance regardless of mean importance. Tasks that fall into the non-relevant category were identified as not meaningful data for the occupational analyst rating process and therefore omitted from data. Presenting the task data in terms of either core or supplementary tasks was considered beneficial, since occupational analysts see the incumbent’s perspective concerning the relative importance of tasks and thus, increase the accuracy of their ratings (Tsacoumis &Willison, 2010).

O*NET is a useful system for investigating task, however some steps in the procedure are debatable. First of all, many tasks are excluded because those tasks are not relevant enough to the job (score of less than 10 %) or are rated by less than 15 people. One of the contributions of this study is to investigate differences between people in the execution of the job, consequently it is important that even tasks that are only done by one or a few people are incorporated.Once those tasks are removed it is harder to give a reliable view about how people perform or see their job in various ways.

Secondly, tasks listed in O*NET are not formulated by the incumbents, they are only rated by them. Therefore, tasks might be excluded, because incumbents had no opportunity to fill in their own job tasks.In this study one of the goals is to explicitly investigate differences between-people and changes over time within-people, as a results even ‘small’ or ‘rare’ tasks can be important. It would be interesting to know, what those ‘rare’ or ‘not important’ tasks are and why people choose to add them to their job.

Finally, when considering the tasks mentioned on O*NET for primary school teachers, it is seen that all the tasks are coded as core tasks, except of two supplementary tasks. This

(14)

makes it hard to analyze changing jobs within-people and variances between-people in different task types. One explanation might be that there are tasks that belong to the category ‘non relevant’ and because those tasks are excluded only many core tasks remain in the list.

In conclusion, O*NET is used in many studies and has some benefits such as an enormous amount of data (e.g. Lievens et al., 2010; Diedorff et al., 2009). Though, in this study different data is used, because tasks that may not be mentioned within O*NET can give insight in how people change their job over time or may give insight how people perform their job differently. Below the theoretical background behind the classification system is used in the present study and the classification itself is described.

Theoretical background task coding

As described above there are several ways to categorize tasks, however in this study none of them are used as those do not comply with the aim of this study. That is, using a classification which can be used to analyse changing tasks and task differences in primary school teachers and can be used to examine other jobs as well. The coding developed in this study is mainly build on the job characteristic model formulated by Hackman and Oldham (1976) and one factor of the knowledge characteristic model formulated by Campion (1988).These models are normally used for the classification of jobs and not for categorizing tasks within a job. In this research this model is used to classify different types of tasks in a job, because they include useful concepts which can also differentiates task types.

Hackman and Oldham (1976) state that there are five job characteristics that makes a job challenging and fulfilling and thereby results in more motivated and satisfied employees. Especially the first three contribute to job meaningfulness and as a results in job satisfaction (Hulin & Judge, 2003). The five job characteristics are described in more detail below.

(15)

thus whether it involves a whole piece of work where the results can be easily identified (Hackman & Oldham, 1976; Sims, Szilagyi, & Keller, 1976). Jobs that involve a complete task are more interesting to perform than jobs that involve only small parts of the task (Hackman & Oldham, 1976).

Task significance is the degree to which one’s work is seen as important and significant. Moreover, it refers to the extent it influences the lives or work of others, both inside and outside the organization (Hackman & Oldham, 1976). In general, jobs that have a significant influence on the physical or psychological well-being of others are likely to provide greater meaningfulness in the work (Hackman & Oldham, 1976).

Skill variety is the extent to which a job allows employees to perform different tasks. Jobs that involve the performance of a number of different tasks are likely to be more

interesting and enjoyable to perform (Sims et al., 1976).

Autonomy is probably the most studied job characteristic, because it has a dominant position in the motivational theories of work (Campion, 1988). Autonomy is the degree to which employees have control and discretion to decide to conduct the job in their own way (Hackman & Oldham, 1976). Later on, studies extended this conceptualization with

autonomy in work schedule, work method and decision making (Morgeson & Humphrey, 2006).

Feedback is degree to which the work itself provides feedback (quality & quantity) concerning how the employee is performing the job and the effectiveness of task performance (Hackman & Oldham, 1976). Thus, it focuses on feedback directly from the job itself or knowledge of one’s own work activities instead of feedback from others (Hackman & Oldham, 1976).

One factor of the knowledge characteristics model is added, because specialization is an important feature in jobs (Campion, 1988). The concept of specialization was first

(16)

recognized by Campion (1988) and later further explained by Edwards, Scully and Brtek (1999). It distinguishes the people who learned the job and thereby have the deep knowledge and skills in that particular area, from people who have little knowledge about that area. It assumes that particular tasks within a job involve performing specialized tasks or possessing specialized knowledge and skills. Another feature in the knowledge characteristic model is the same as the one in the job characteristics model, namely skill variety. The knowledge characteristics model reflects the kinds of knowledge, skills and abilities needed to get the job done. The remaining knowledge characteristics are job complexity and information

processing (Morgeson & Humphrey, 2006). Since taking job complexity and information processing into account makes the coding too lengthy and does not add much more, those are not used.

Task coding

On the basis of the above described job/knowledge characteristics, three types of tasks are formed. Those categories are core, noncore and extra tasks and are explained more

extensively below.

Core task, are tasks that are fundamental, uniquely, and specific to that kind of job.

Tsacoumis and Willison (2010) describe core tasks in O*NET as tasks that are critical to the jobs as is the case for this study. Important factors in people’s job are job meaningfulness and job satisfaction, for example Suadican, Bonde, Olesen and Gyntelberg (2013) found that intentions to quit are influenced by low job satisfaction and low job meaningfulness. As mentioned before, as a job scores high on skill variety, task significance, task identity,

autonomy and feedback, job satisfaction and job meaningfulness tend to be high (Hackman & Oldham, 1976). Therefore, it is suggested that core tasks score high on those job

(17)

choose that specific job. As O*NET emphasizes that task importance is essential for their ‘core task’ type, this would also be interesting for this research. Gabe and Abel (2012) related task importance rated by O*NET to specialized knowledge, which refers to the distinctiveness of the cognitive skills required in a job. Therefore, it is suggested that core tasks also score high on the knowledge characteristic specialization, because to perform well on core tasks someone needs a certain amount of depth and knowledge in that particular area (Gabe & Abel, 2012). Examples of core tasks within the teaching job are teaching and preparation.

Non-core tasks are tasks that are important to the job, but are not specific fundamental and uniquely for that job. O*NET included tasks such as administration and reports as core tasks for teachers, because they are important to the job (onetonline.org). However, in this study those tasks are not included in core tasks, since tasks as administration and meetings are not uniquely important to the teaching job. In sum, noncore tasks are important to the jobs, but are also important in other jobs and are not job specific. Noncore tasks score low on task significance and specialization and might score high on all the other characteristics. Noncore tasks score low on task significance, because it does not contribute to meaningful work since the tasks does not influence the physical or psychological well being of others (Hackman & Oldham, 1976). Moreover, noncore tasks score low on specialization because those tasks are also done in other jobs thus no specific knowledge is needed for those tasks. Examples of noncore tasks are administration, general meetings and self development in the form of study days.

Extra tasks are tasks that only some employees perform, thus are not included in their job description. Extra tasks might be uniquely, but not fundamental to the job. In O*NET a task must be rated by more than 15 people to be included and must score high on relevance (Tsacoumis & Willison, 2010). Consequenlty, some extra tasks included in this study might not be mentioned in O*NET, because only one or a few people mentioned those tasks or those

(18)

tasks were not relevant enough to the job. Extra tasks might score high on task identity, task significance, skill variety, autonomy and specialization. However, these tasks are not done by every employee in the same job and therefore cannot include feedback. As Hackman and Oldham (1976) stated feedback is directly coming from the job and concerns how the employee is performing the job. Since, an extra task not provides feedback about the job specific performance this does not includes feedback. Examples of extra tasks are doing commitees or being a coordinator.

Hypothesis

First the hypotheses about changing jobs within-people are discussed followed by hypothesis about the differences between variations within-people and between-people. Lastly, the hypotheses about job meaningfulness and job satisfaction are discussed.

Firstly, core tasks are fundamental to the job (Tsacoumis and Willison, 2010),

therefore core tasks cannot be done less or cannot be disposed. This is also the case for adding core tasks, because those tasks are critical to the job, it is suggested that extra tasks are done from the beginning. Moreover, job crafting can play a role in changing jobs and differences between job ratings (Lievens et al., 2009). Since, core tasks score high on all the five job characteristics that lead to job meaningfulness and job satisfaction it is unlikely that

employees try to achieve a better person-job fit through crafing core tasks or are motivated to craft their core tasks (Hackman & Oldham, 1976; Tims & Bakker, 2010).

Secondly, as for core tasks, noncore tasks are important to the job, they should be done in order to perform well on the job. For example, one should do the administration otherwise problems might occur. However, noncore tasks are not as fundamental to the job as core task. Changes in noncore task might occur because of the increasing technological innovations, which makes it easier to communicate with each other and to have virtual

(19)

meetings instead of face-to-face (Cascio, 1995; Burke & Ng, 2006). Furthermore,

technological innovations can influence how administration is done, it might be easier to get it done.

Thirdly, acquiring a better person-job fit is one of the reasons why people make changes in their job (Tims & Bakker, 2010). Person-job fit is often seen as a complementary fit. Muchinsky and Monahon (1987) operationalized complementary fit as an individual’s skills meeting environmental needs, in other words demands-ability fit. Kristof (1996)

expanded this by including that individuals needs are met by environmental supplies which is called needs-supply fit. In case of a misfit between the needs of the individual and the

environmental supplies, employees can make changes in their job. However, the job demands employees to fulfill certain tasks that are fundamental to the job. In line with this research it can be suggested that employees decrease the misfit between their needs and the

organization’s supplies by changing tasks that are not fundamental to the job. In this study this would mean that extra tasks are crafted more extensively than core and noncore tasks. On the basis of the above mentioned arguments the following hypothesis are formulated:

H1a: Core tasks are the least open to job changes in compared to non-core and extra tasks

H1b: Noncore tasks are more open to job changes compared to core tasks, but less open to job changes compared to extra tasks.

H1c: Extra tasks are the most open to job changes compared to non-core and core tasks

It is expected that more differences exists between-people than within-people as is explained below. Between-people refers to differences in how different people perform their job and within-people refers to changes one person makes in their job over time.

(20)

As earlier mentioned in the theoretical background employees are unique in how they define and shape a job (Morgeson & Diedorff, 2011; Lievens et al., 2009; Sanchez & Levine, 2012). There are other factors besides individual motivation which influences the shaping and interpretation of tasks. For example, familiarity/job tenure and general cognitive ability (Morgeson & Diedorff, 2011). Those factors might have an influence on how people see their tasks. Individual differences are likely to influence all kind of tasks. For example, an teacher which already has ten years experience spends most likely less time on correcting and preparation than someone who just started.

In addition, job crafting might be more visible in one person than another person. Job crafting is a form of proactive behavior. These have in common that the person initiates them either by acting in advance on a future situation and/or by taking control and causing change (Parker & Collins, 2010). Job crafting is different from previously studied proactive

constructs, because the changes that job crafters make are primarily focused on improving their person–job fit and work motivation (Tims & Bakker, 2010). Some people are more proactive and are seeking for more challenges than other people. Employees with a proactive personality are more prone to show initiative, take action, and persist until they bring about meaningful change than employees without this personality characteristic (Crant, 2000). It is expected that people with more proactive personality characteristics hold more tasks and seek for more challenges. However, when someone is less proactive it is suggested that less

initiative is shown in the whole career. Therefore, it is expected that between people there are more differences in core tasks, noncore tasks and extra tasks.

H2: Differences on core, noncore and extra tasks are more seen between-people than within-people

In the development of the categories core, noncore and extra tasks, it is suggested that core tasks lead to more job meaningfulness and job satisfaction which is based on the job

(21)

characteristic model of Hackman and Oldham (1976). On the basis of the reasoning behind core tasks it is could be suggested that teachers that spend more time on core tasks are more satisfied with their job and think their job is more meaningful. To test this reasoning behing core tasks, two additional hypothesis are test.

H3a: Teachers spending much time on core tasks are more satisfied with their job than teachers spending more time on core tasks.

H3b: Teachers spending much time on core tasks think their job is more meaningful than teachers spending less time on core tasks.

In sum, the theoretical framework has shown how job analysis is seen in a different light because of external changes, job crafting and differences between people. Moreover, the job analysis and job characteristic literature is used to form three types of task: core tasks,

noncore tasks and extra tasks to study job changes more extensively. The research question in this first study is: ‘What types of tasks are more open to job changes?’, since not much

research is done on the specific tasks that change within jobs. It is expected that within-people extra tasks are the most open to job change. Furthermore, it is proposed that there are more differenes between than within-people in jobs.

 

Research  method      

The first study is descriptive and is designed to produce an accurate presentation of tasks being performed by primary school teachers. It is a deductive approach since hypotheses are based on the existing literature and tested by analysing results from the data gathered

(Saunders & Lewis, 2012). Through the data from 134 elementary school teachers it can be studied whether there are differences between-people. In addition, participants answer

(22)

questions in which it becomes clear whether a task is new and whether the time spend on the tasks has increased/decreased. The purpose of this study is to get a better insight in what type of tasks are changing and this study allows to get information about the time spend on tasks and the amount of tasks in the beginning of the job and now. Another advantage of this study is that people have to come up with their own opinions of the tasks that are included in their job, therefore we get a good understanding how people see their jobs.

Mooi Werk Tool

The quantitative data is obtained by using a web-based application tool. It is called ‘Mooi Werk Tool’ and designed by TNO and the University of Twente. The application tool includes a stepwise process in which the respondents have to answer several questions and have to do some exercises. The tool starts with some demographic questions and work related questions. An example of a demographic question is ‘In which year are you born’? and an example of a work related question is ‘How long are you working for your current

employer?’. The measures of job satisfaction and job meaningfulness are both gathered in the first part where the demographic and work related questions are asked.

Job satisfaction is studied by asking the question ‘To what extent are you, all things considered, satisfied with your work?’ and is measured by a 5 point Likert scale ranging from ‘very unsatisfied’ to ‘very satisfied’. Job meaningfulness is studied by asking the question ‘To what extent do you think your work is meaningful?’ and is measured by a 4 point Likert scale ranging from ‘never’ to ‘always’.

Seven steps are included in the tool. This study uses only the first three steps, because those steps can test the formulated hypotheses. The three steps are visualised in order to give the tooluser a better understanding what has changed in their jobs.

(23)

1. Task diagnoses: identification of the tasks that people perform in their job. The question asked is: ‘What are the individual tasks/work activities included in your job at the moment?’ Users of the tool distinguish the different tasks in their job. Toolusers write down those tasks by themselves and the tasks are not necessarily listed in their job description.

2. Task diagram: on the basis of the tasks mentioned in step 1 by the tooluser, the tasks are divided into three segments. Toolusers allocate their tasks into three different segments: small tasks, moderate tasks and large tasks. Furthermore, toolusers report for each segment how much percent of their work time they devote to the tasks in that specific segment.

3. Task dynamic: toolusers are challenged to think about how their job has changed over the years. The purpose of this step is to get an understanding of how the task dynamics are in terms of whether new tasks are added or whether tasks have shrunked or growth over time. On the basis of the earlier distinction made between tasks, toolusers have to divide those earlier listed tasks in the following categories. Firstly, they have to choose whether a ‘task is done from the beginning’ or whether a ‘task is added later’. Secondly, they have to select whether the ‘time spend on the task is shrunked over time’ or whether the ‘time spend on the task has grown over time’. The results of this step are visualised in a crosschart, with the horizontal axes running from ‘done from beginning’ to ‘added later’ and the vertical axes running from ‘has shrunked over time’ to ‘has grown over time’.

Sample

The data from the ‘Mooi Werk Tool’ is secondary, because the data was originally gathered for the purpose of TNO and the University of Twente to investigate job crafting. The data is gathered in a workshop setting, with an instructor guiding every team of primary school teachers through the steps of the tool. A total of 134 teachers from nine different primary schools completed the ‘Mooi Werk Tool’ between January and March 2015. The schools are

(24)

all located in and around Enkhuizen, Noord-Holland and are all part of Stichting Katholiek Onderwijs (SKO), which is a foundation for Catholic primary education.

A total of 19 respondents were excluded from the study, 6 subjects did not teach and 13 subjects did not fill in the demographic data. One respondent that only missed age as demographic variable was included. No respondents were excluded because of outliers, because the intention of this study is to investigate the differences. The remaining 115 primary school teachers included 98 female and 17 male. This is not a problem, since the sample is representive for the population. Between 1999 and 2002 the percentage female is approxamately 80% in primary education and the prospect in 2004 was that it would increase slightly (Driesen & Doesborgh, 2004) Based on 114 respondents, the age is between 25 and 65 (M=46, SD=11,1). 106 teachers hold a HBO diploma, the remaining 9 teachers hold a HAVO/VWO, MBO, or WO diploma. The average tenure in the function is 6 years (SD=8.91) and the average amount of working hours is 29 (SD=9.61).

Procedure

The tasks identified by the toolusers during the first step of the tool, ‘task diagnoses’, are coded as core, noncore and extra task by using a self developed classification system. As explained in the theoretical background, this coding system is build on the job characteristics theory and the job knowledge theory (Campion, 1988; Hackman & Oldham 1976). Two researchers (me and another Master student in Business Administration) came together several times to discuss how to divide the tasks into core, noncore and extra task types. During the coding we kept close contact, when something was unclear we discussed in which category the task belonged. The researchers made a table (Appendix 1) to assign every task they saw to one of the task types. Below some examples of tasks in the different task types are given:

(25)

• Core tasks: teaching, correcting and preparing • Noncore tasks: meetings and administration • Extra tasks: comittees and reading coordinator

Since, teachers had to identify the tasks by themselves, some tasks were ambigious and hard to assign to a specific task type. Particularly, when people mentioned several tasks as one task, whereby one task belongs in another category than the other task. For example, some people have meetings and comittees included as one task. However, it was classified as respectively noncore and extra tasks. It was decided that those should be appointed to noncore tasks, although not all teachers do committees, meetings however are a part of the job.

Another example is that some teachers named a task ‘tasks’, it is not clear what they meant with this. ‘tasks’ is assigned to extra task type, because only a very small amount of teachers mentioned it.

For each type of tasks there are five measures used. Time spend on tasks: how much of the work time someone spends on that type of task. Amount of tasks: how much tasks someone has in one of the task types. Old tasks: tasks that are done in the beginning of their job. Tasks: tasks that are done now. New tasks: tasks that have been added.

Data analysis

Firstly, job changes within-people are tested. Hypothesis 1a states that ‘core tasks are the least open to job changes compared to non-core and extra tasks’, hypotheses 1b states ‘noncore tasks are more open to job changes compared to core tasks, but less open to job changes compared to extra tasks’ and hypothesis 2c states ‘extra tasks are more open to job changes compared to non-core and core tasks’. All three hypothesis are about differences in tasks between two points in time within-people. More specifically, the difference between amount of tasks at the beginning of the job (time 1) and amount of tasks in the job currenlty

(26)

(time 2) are tested. Moreover, the differences between time spend on tasks at the beginning of the job (time 1) and time spend on tasks in the job currently (time 2) are tested. The

differences within-people are tested using a reliability measure to investigate whether there is a high consistency between the amount of tasks and the time spend on tasks in the beginning and now. When there is a high consistency this would suggest that not much has changed in the jobs of people. The specific reliability measure used is described below.

On the basis of the paper from Rankin and Stokes (1998) the intraclass correlation (ICC) is used for testing the reliability within the teachers. In this paper it is discussed why other measures such as the Pearson’s correlation coefficient and a paired t-test are

inappropriate measures for reliability. The ratings for the ICC can be from two people (or to types of equipment), or the same person on two, or more occasions when the variables are continuous (Kottner et al., 2011; McGraw & Wong, 1996; Rankin & Stokes, 1998). Also referred to as test-retest reliability/agreement. Results of the ICC provide information about the amount of error inherent in the measurement. Confidence intervals as measures of

statistical uncertainty should be reported, because the range of values that are considered to be plausible for the population of interest are useful for interpreting results (Kottner et al., 2011). The results of agreement are classified as follows: sligh agreement, 0.00-0.20; fair agreement agreement, 0.21-0.40; moderate agreement, 0.41-0.60; sustantial agreement 0.61-0.80 and almost perfect agreement, 081-1.00 (Kottner, et al., 2011).

Secondly, a comparison is made between the difference between-people and the differences within-people. Hypothesis 2 states that ‘differences on core, noncore and extra tasks are more seen between-people than within-people’. To test whether there are more differences between people than within people Levene’s test for equality of variances is used. It tests whether the variance in one variable is equal to the variance in another variable (Lim & Loh, 1996). More specifically, in this study it tests whether the variance between

(27)

within-people and between-within-people in core, noncore and extra tasks is equal. The variance within people is rated by taking the variance in the ‘new task’ variables, because these are the difference between old tasks and current tasks.

Variance represents the average squared deviations between a group of observations and their specific mean, also called homogeneity of variance (O’Neill & Mathews, 2000). It is a measure of spread. Homogeneity of variance is a common assumption across several

statistical analysis (e.g. t-test, Anova). Though, it can be also useful for testing hypotheses as well. Traditional Levine F test measures the mean difference between absolute differences between each observation and its corresponding mean. Problem with this test is that it is sensitive to variations in the distribution. Thus, when it is not normally distributed, it tends not to be powerful enough. The assumptions to use this test are: to have normally distributed data or more than 100 observations and to have equality in sample size between the

groups(Lim & Loh, 1996). These assumptions are both met.

Thirdly, in the theoretical background it is assumed that core tasks score high on the 5 job characteristics formulated by Hackman and Oldham (1976). According to them jobs high on those characteristics result in higher job satisfaction and job meaningfulness. It is

hypothesed (3a) that ‘teachers spending much time on core tasks are more satisfied with their job in comparison to teachers spending more time on core tasks.’. Furthermore, it is

hypothesed (3b) that ‘teachers spending much time on core tasks think their job is more meaningful in comparison to teachers spending less time on core tasks.’ This is tested by using scatterplots to see whether there is a relationship between time spend on core tasks and job meaningfulness/satisfaction. Furthermore, the Spearman’s rho is used for correlations, because job satisfaction and job meaningfulness are measured on a nominal scale.

(28)

Results

Descriptive statistics

The descriptive statistics are shown to get an indication whether there are differences within and between-people in core, noncore and extra tasks. ‘New’ refers to tasks that are added to the job later, ‘old’ refers to tasks that are done in the beginning of the job. Below in Table 1, 2 and 3 under ‘core new, noncore new and extra new’ it can be seen that the amount of core (M= 0.78, SD= 1.06) and noncore (M= 1.02, SD= 0.96) tasks is less changed over time than the amount of extra tasks (M= 1.75, SD= 1.58). In addition, this is also the case for the time spend on core (M= 7.03, SD= 12.76) and noncore tasks (M= 7.42, SD= 8.93) in comparison to the time spend on extra tasks (M= 10.43, SD= 1.82).

Tabel 1: Descriptive statistics of core task amount and core task time

Mean SD Variance Median Minimum Maximum

Core amount 4.68 1.37 1.89 5 1 8

Core time 61 14.64 214.33 62 23 91

Core new amount 0.78 1.06 1.12 0 0 6

Core new time 7.03 12.76 162.96 0 0 78

Core old amount 3.91 1.26 1.59 4 1 8

Core old time 53.55 17.17 294.92 55 3 86

Tabel 2: Descriptive statistics of noncore task amount and noncore task time

Mean SD Variance Median Minimum Maximum

Noncore amount 3.43 1.46 2.14 3 0 7

Noncore time 24.23 13.02 169.60 22 0 55

Noncore new amount 1.02 0.96 0.930 1 0 4

Noncore new time 7.42 8.93 79.70 5 0 38

(29)

Noncore old time 16.81 13.05 170.28 14 0 6

Tabel 3: Descriptive statistics of extra task amount and extra task time

Mean SD Variance Median Minimum Maximum

Extra amount 2.86 1.80 3.23 3 0 8

Extra time 15.91 13.78 189.82 12 0 76

Extra new amount 1.75 1.58 2.49 1 0 6

Extra new time 10.43 11.82 139.69 6 0 50

Extra old amount 1.10 1.13 1.29 1 0 5

Extra old time 5.50 6.85 46.00 3 0 43

Intraclass correlation to measure within subject differences

The descriptive data shows that more changes are seen within-subjects in extra tasks compared to noncore and core tasks, however this is tested more deeply by using the ICC. Hypothesis 1a, 1b and 1c are tested using the intraclass correlation. It is tested whether the amount of tasks and time spend on tasks is consistent at time 1 (beginning job) in comparision with time 2 (currently). This is done for core, noncore and extra tasks.

Table 4: Intraclass correlation and the 95% confidence interval for the reliability between tasks done from the beginning and tasks done currently.

ICC single 95% CI

All tasks amount -0.245 -0.409 - -0.66 Core task amount 0.543 0.400 - 0.659 Core task time 0.583 0.448 - 0.691 Extra task amount 0.079 -0.104 - 0.258 Extra task time 0.149 -0.034 - 0.322

(30)

Noncore task amount 0.595 0.463 - 0.709 Noncore task time 0.635 0.512 - 0.732

Overall, Table 4 shows that there is a poor agreement between the amount of tasks done at the start of the job and tasks done at the moment (ICC= .245). When looking at the task types seperately, it can be seen that there is moderate agreement between the amount of core tasks done at the beginning of the job and done currently (ICC= .543), as is the case for the time spend on core tasks (ICC= .583). In addition, the agreement between the amount of noncore tasks is also moderate (ICC= .595), as for the time spend on noncore tasks (ICC= .635). In contrast, the agreement between the amount of extra tasks done at the beginning of the job and done currently is poor (ICC= .079), as is the case for time spend on extra tasks (ICC= .149).

Hypothesis 1a and 1b are partially in line with the results, it was not expected that noncore tasks would be equally open to job changes as core tasks. Though, it was expected that both core and noncore tasks showed a higher consistency between time 1 and time 2 in than extra tasks, both amount and time spend on tasks. Thus, hypothesis 1c is in line with the results.

Levene’s F test for homogeinity in variances

To test hypothesis 2, Levine’s test for homogeneity in variances is used. Firstly, the results show that indeed differences on core tasks and noncore tasks are more seen between people than within people. The variance between people on core task time (σ2=214.33) is significant higher than the variance on core task time within people (σ2=167.44), which means that there are significant more differences in variances between-people than within-people on core task time (F(1, 228)= 10.36, p= .001). The same holds for core task amount, the variance is

(31)

significantly higher in core task amount between people (σ2= 1.89) than within people (σ2= 1.12) (F (1, 228)= 4.68, p= .0.32). For noncore tasks only the variances on task time between (σ2= 169.60) and within (σ2= 79.633) people differ significantly (F (1, 228)= 10.938, p= .001). Only the variance in noncore task amount is not significantly different between within-subjects (σ2= 0.93) and between-subjects (σ2= 2.14). However, the variances are nearly significant (F (1, 228)= 3.23, p= .073), with higher variances between people.

Secondly, the differences in extra tasks are both seen within and between people. The variances between-people and within-people do not differ significantly from each other. Thus, hypothesis 2 is partially in line with the results. Both between and within-subject show big variances for extra task time and extra task amount. It is known from the ICC that the

consistency of extra tasks within people are significant different from each other. The results show that the variances in extra task amount between within-subjects (σ2= 2.50) and between-subjects (σ2= 3.23) do not differ significantly F (1, 228)= 3.45, p= .065). In addition, the variances in extra task time between within-subjects (σ2= 189.82) and between-subjects (σ2= 139.58) do not differ significantly from each other (F (1, 228)= 0.03, p= .856). This indicates that when within-subjects the consistency is low, the consistency between-subjects is also low.

In sum, hypothesis 2 is partially in line with the results, only differences in core and noncore tasks are more seen between-people than within-people. For extra task the variances between within-people and between-people are both large.

Core tasks and job meaningfulness/satisfaction

To test the following hypothesis, scatterplots are made and correlations are calculated. ’Teachers spending much time on core tasks are more satisfied with their job in comparison to teachers spending more time on core tasks.’ and ‘teachers spending much time on core

(32)

tasks think their job is more meaningful in comparison to teachers spending less time on core tasks.’ No further analysis were done, since the results showed that there was no relationship between core task time and job meaningfulness/job satisfaction. As can be seen in the

scatterplots below (Figure 1-2), there is no relationship between core task time and job meaningfulness/satisfaction. In addition, Spearman’s Rho indicates that there is a very weak relationship between core task time and job satisfaction/meaningfulness (Table 5). It

implicates that the reasoning behind core tasks is not in line with these results.

Figure 1: scatterplot showing the relationship between core task time and job meaningfulness

Figure 2: scatterplot showing the relationship between core task time and job satisfaction.

(33)

Table 5: Relationship between core task time and job satisfaction/job meaningfulness calculated by Spearman’s rho

N Spearman’s rho Sign (2-tailed) Core task amount * job satisfaction 113 -0.140 .140

Core task time * job satisfaction 113 -0.103 .279 Core task amount * job meaningfulness 115 0.137 .146 Core task time * job meaningfulness 115 0.064 .498

Conclusion

The focus of this first study was to investigate whether people perform their job different in the beginning than currently and whether there are many differences between-people in how they execute their job. It is studied on a task level to indicate whether some task types are more open to changes/differences than other task types. The results showed that extra tasks are more open to changes within-people than core and noncore tasks. Thus, time spend on extra tasks and the amount of extra tasks are much different in the beginning of the job compared to the job at the moment. People are more consistent in core tasks and noncore tasks over the years.

When investigating whether there are more differences between-people than within-people in task amount and task time it is seen that there are more differences between-within-people compared to within-people in core and noncore tasks. This indicates that noncore and core tasks maintain relatively stable within people, but differ between people. The differences in extra tasks between-people and within-people are both large and about the same.

Link Study 1 and Study 2

According to Saunders and Lewin (2012) one of the reasons to use a mixed model

(34)

part gives answers to the question whether a job changes over time and whether there are differences between how people perform the same job. Furthermore, it shows whether those differences/changes vary within different task types. However, this gives no information about why jobs are changing, whether people make changes because of their own initiative or because of external factors.

Another reason to employ a mixed model is using the qualitative methods to explain relationship between quantitative variables (Saunders & Lewis, 2012) . This is also the case for this study. Results showed that extra tasks are more open to job changes within-people, it is interesting to investigate this more during an interview. Furthermore, it is interesting to investigate whether different forces behind job changes influences the distinctive task types. In sum, the purpose of the interviews conducted in the second part of this study are to get an understanding why tasks are changing and why extra tasks are more open to job changes in comparison to core and noncore tasks.

Study 2: Forces behind job crafting Literature review

A second goal of this study is to investigate the forces behind changing jobs. Studies identified situational or individual factors that influence job changes (Sanchez, 1994; Wresniewski &Dutton, 2001). One important individual factor of changing jobs is job crafting, which is discussed below (Tims, Bakker & Derks, 2012). Furthermore, situational factors that can result in job changes are discussed.

Job crafting

The fundamental concept of job crafting is that employees alter their tasks or other job characteristics upon their own initiative (Tims, Bakker & Derks, 2012). Job crafting is an

(35)

important principle within the organizational literature, since it is associated with several positive outcomes (Demerouti, 2014). Examples of these outcomes are an increase in motivation, work engagement, experienced meaning, job satisfaction, health and job performances (Bakker, Demerouti & Euwema, 2005; Berg, Grant & Johnson, 2010; Lyons, 2008; Nielsen & Abilgaard, 2013; Petrou et al. 2012; Tims et al., 2012; Tims, Bakker & Derks, 2013).

Job crafting is conceptualized from two different perspectives. One of them is the view from Wrzesniewski and Dutton (Demerouti, 2014). They define job crafting as “the physical and cognitive changes individuals make in the task or relational boundaries of their work” (p. 179). According to this perspective employees redefine jobs to their own motives, strenghts and passions (Wrzesniewski, Berg & Dutton, 2010). Employees achieve this by changing their task, cognitive task or/and relational boundaries. Task boundaries focus on the form, scope and the number of tasks one is involved in while working. Altering task

characteristics is achieved by choosing to do fewer, more, or different tasks than prescribed in the formal job. Cognitive boundaries emphasises how someone sees the task. Crafting

cognitive boundaries involves altering the view of work as a whole or the different aspects of a job. Relational boundaries focus on the relation with supervisors and other employees and can be altered by changing with whom one interacts at work and the nature of interactions at work (Demerouti, 2014). Job crafting can occur in any type of job and the purpose is

increasing meaning in work and changing the identity and the role in the organization (Wrzesniewski & Dutton, 2001) .

A different view on job crafting focuses on job demands and job resources. The job demands-resources model (JD-R model) emphasizes that job crafting is done to balance between job demands, job resources and employees’ personal needs and abilities (Petrou et al., 2012; Tims & Bakker, 2010). Job demands refer to the aspects of jobs that require

(36)

sustained physical and/or psychological (cognitive and emotional) effort or skills and are therefore associated with certain physiological and/or psychological costs (Bakker &

Demerouti, 2007). Job resources are either/or functional in achieving work goals, reduce job demands and the associated physiological and psychological costs, and stimulate personal growth, learning, and development. Both demands and resources can be physical,

psychological, social, or organizational (Bakker & Demerouti, 2007).

Tims et al., (2012) validated four dimensions of job crafting in this JD-R model: increasing structural job resources; increasing social job resources; increasing challenging job demands; decreasing hindering job demands. Increasing structural job resources refer to variety in resources, opportunity for development, and autonomy and are likely to have more effect on job design, because it is about gaining more responsibility and or knowledge about the job. Increasing social job resources refers to social support, supervisory coaching, and feedback and may have more impact on the social aspects of the job and attaining satisfactory levels of interaction (Tim et al., 2012). Increasing the level of challenging job demands means that people might seek for challenges to increase their personal growth and satisfaction with the job (Berg, Duton & Wresniewski, 2008). A job that is under stimulating can result in boredom and in turn can cause absenteeism and job dissatisfaction (Kass,Vodanovich & Callender, 2001).   People may make changes to their job to attain the job interesting and motivating and therefore the employee can become more engaged in to the job (Crawford, LePine & Rich, 2010). Lastly, people might make changes to a job in order to decrease the level of hindering job demands. Prolonged exposure to high job demands with a low level of job resources can result in several negative outcomes, such as a burn out (Bakker et al., 2005).

Both Wrzesniewski and Dutton’s perspective and the JD-R model agree that

employees change their job/tasks in order to manage their problems better and find solutions for these difficulties. In addition, both perspectives agree that also in stable work

(37)

environments in which work procedures are provided with clear job descriptions jobs are adapted by employees to their own needs (Petrou et al., 2012; Werzsniewski & Dutton, 2001). However, the way employees redesign their jobs to tackle those problems/difficulties differ in the two perspectives as discussed below (Demerouti, 2014).

Wrzesniewski and Dutton (2001) focus on task oriented and relational oriented job crafting. Their suggestions about task oriented job crafting are changing, the number, scope and type of job tasks and do not specify what types of tasks are changing. More specifically, Wrzesniewski and Dutton (2001) explain how tasks can be changed and give examples of those changes (e.g. “Design engineers engaging in relational tasks that move a project to completion”), but do not provide what types of tasks are changing within those occupations. As Wresniewski and Dutton, Lyons (2008) investigated job crafting in sales representatives in a general way, the categories of job crafting he used were: personal skill development (e.g. “Deciding to learn basic Spanish”), task function (e.g. “Expanding demo material on hand”), advancing relationships (“Visit more persons on site visits”), tactics choices (“Create reading program of books, magazines etc. to locate novel sales methods”) and maintaining

relationships (“Guarantee contact with actual purchaser”).

Job crafting in the JD-R model can be described as proactively seeking to enhance resources and challenges and reducing demands (Petrou et al., 2012; Tims & Bakker, 2010). This view focuses on general aspects in work that change as can be seen in the job crafting scale from Tims et al. (2012). Examples of statements in this scale are: “When there is not much to do at work, I see it as a chance to start new projects” and “I decide on my own how I do things”, however those imply nothing about the type of tasks that are changing.

As above explained the two perspectives on job crafting are still general and do not provide an understanding of what specific tasks are crafted and to what extend these tasks are crafted. The JD-R model and Wresnieuwki and Dutton only explain how jobs can be changed

(38)

(e.g. respectively reducing demands and changing number of tasks) and give general

examples of changing tasks, it is not specified on task types. Lyons (2008), divide job crafting behavior in several types, but again those behaviors are not related to specific task types within jobs. The first part of this study showed that changes are mostly seen in extra tasks in comparison with core and noncore tasks. Extra tasks are only performed by some employees and are not fundamental to a job, they are not included in the job description. Besides extra tasks, also changes are made in core and noncore tasks, however it is done less. It would be interesting to investigate in more detail why extra tasks are more open to job changes than core and noncore tasks and whether this is due to job crafting (internal motivations) or to external forces.

External forces

Several external forces can lead to changing jobs. In study 1 some external forces are mentioned such as globalization, technological innovations, shift from a manufacturing to a service based economy and changes within the organization or relationships within the organizations (Burke &Ng, 2006; Cascio, 1995; Sanchez, 1994).

Reorganizations and mergers also involves job changes for employees. Task

assignments and geographical locations are changed in order to meet the changing demands that result from the rapid economical and technical developments (Dam, 2005). In another study it was found that employees crafted their jobs, because a merger resulted in

misalignments between the employees’ identity and the work carried out (Kira, Balkin & San, 2012).

In summary, much more can be found about internal forces of job changes. There are external forces that influences job changes, however most external forces discussed in the literature are

Referenties

GERELATEERDE DOCUMENTEN

Department of Obstetrics and Gynecology University of Mississippi Medical Center United States of America..

As the society and the industrial ecology community grew, the ISIE created sections focused on industrial symbiosis/eco- industrial development, socioeconomic metabolism,

The following subjects are discussed during the interviews: the process concerning choosing the appropriate study, more specific the wants and needs of people concerning

In this study we will address certain aspects that are important to generate proper results. It will give a visual on how firms choose certain strategies and how they move

could be included in Flyvbjerg‟s analysis and it is closely related to the Venetian case study and to the urgency of a structure that would protect the lagoon. The motivation

ANDANTEK differentieels serie SR kunnen worden gebruikt voor een groot aantal toepassingen.. Enkele voorbeelden zijn hieronder

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of

The top 3 innovations that are the least likely to get adopted by an intuitive individual, are creative job titles (16), health risk assessment (41) and AI video analysis (77),