Flexible work designs, well-being and the role of personality

(1)

Master Thesis

Flexible Work Designs, Well-Being and the Role of Personality. Eric de Vries

University of Amsterdam

Name: Eric de Vries Student ID: 5883857

Master: Corporate Communication Supervisor: Claartje ter Hoeven

(2)

1 Abstract

Flexible work design have gained popularity in organisations and new communication technology has facilitated this process. However, studies on flexible work designs have yielded contradicting results. In this study, through the exploration of paradoxes related to flexible work designs, three resources and three demands associated with flexible work designs, were investigated. The focus of this study was on how flexible work designs relates to the well-being of employees, and what role personality plays in this relationship. The findings indicate that flexibility over workplace positively influenced an employee’s well-being, through job autonomy and work-life balance. Communication technology use had a positive effect on an employee’s well-being, through job autonomy and effective and efficient communication, and a negative influence on well-being through interruptions and work intensification. Except for neuroticism, all personality dimensions had some moderating effect on the relationships between the flexible work designs and the flexible work design resources or demands experienced. Theoretical and practical implication are put forth, as well as direction for further research.

(3)

2 Flexible Work Designs, Contradictions and Personality

Flexible work designs (FWDs) have gained popularity in organizations (Kersley, Forth, Bryson, Bewley, Dix, & Oxenbridge, 2006) and new communication technology has facilitated this process. It could be said that especially because of new communication technologies, work processes are able to become more flexible. With FWDs employees gain more autonomy over where and when to work, and which media technology will be used to maintain communication with relevant others (Ten Brummelhuis, Bakker, Hetland, &

Keulemans, 2012). While some organizational benefits of FWDs are clear (e.g., reduced costs for office space; Di Martino & Wirth, 1990; Kurland & Bailey, 1999), other more individual work outcomes (e.g., individual productivity, stress, job satisfaction) are contradicted in research (Allen, Johnson, Kiburz, & Shockley, 2013; De Menezes & Kelliher, 2011; Golden, 2009; Putnam, Myers, & Gailliard, 2013).

A systematic review concerning the relationship between flexible work arrangements and several work outcomes shows that in one-third of the studies well-being is enhanced (De Menezes & Kelliher, 2011). The remaining two-thirds of the studies show that there is no association between flexible working arrangements and an employee’s health or well-being. The differences in findings might be due to certain conditions or processes that facilitate or negate the relationship between FWDs and well-being (De Menezes & Kelliher, 2011). For example, McNall, Masuda and Nicklin (2009) found that flexible work arrangements helped employees to experience enrichment of their private lives, which in turn was associated with higher job satisfaction and lower turnover intentions. Also, the use of communication technology is key to FWDs and Day, Paquet, Scott and Hambley (2012) found that certain ICT demands (e.g., employee monitoring) and ICT resources (e.g., personal assistance) are associated with the use of communication technology. While the ICT demands were

(4)

3 with lower stress, strain and burnout (Day et al., 2012). These studies give some insight that there are underlying processes and conditions which might be key to understanding the contradicting findings of FWDs and the well-being of employees.

To gain perspective into the workings of processes and conditions that are associated with FWDs, in this research the job demands-resources (JD-R) framework will be used (Demerouti, Bakker, Nachreiner, & Schaufali, 2001). Aspects of a job that require sustained physical and/or psychological effort are seen as job demands (Demerouti et al., 2001). Job resources on the other hand, are aspects of a job that could facilitate achieving work goals and reduce the possible strain of job demands (Demerouti et al., 2001). An advantage of using the job demands-resources framework is that contradictions that are associated with FWDs are put in perspective of each other. When clarifying contradictions that exist, a clearer insight is gained into FWD research and its effects on employees’ well-being.

Certain contradictions that are associated with FWDs are described in several paradoxes. The idea of a paradox, where FWDs intended to create job resources for employees also create unintended job demands through the actions of employees or co-workers, will be central in this research. The FWD demands and resources which will be investigated in this research, will be extracted from the described paradoxes.

One of those paradoxes is put forth by Putnam, Myers, and Gailliard (2013) and is called the autonomy-control paradox. This paradox is similar to the autonomy paradox described by Mazmanian, Orlikowski, and Yates (2013). It entails a perpetuating process in which employees gain autonomy because of flexibility over workplace, but at the same time this autonomy is challenged because of the constant interactions and engagement with co-workers (Mazmanian et al., 2013; Putnam et al., 2013) In the (self-perpetuating) paradox of

(5)

4 more organized in their work, but at the same time this technology also creates

disorganization in their work (Rennecker & Godwin, 2005). Ten Brummelhuis et al. (2012), for example, found that FWDs create more effective and efficient communication. On the other hand, communication technology use can increase the amount of interruptions during work, when working on an alternative location (Fonner & Roloff, 2012). Other paradoxes which will be explored are the connectivity paradox (Leonardi, Treem, & Jackson, 2010) and

the telecommuting paradox (Gajendran & Harrisson, 2007).

These paradoxes will be described in more detail later on, but from the short

descriptions given it becomes clear that individuals gain certain resources from FWDs (i.e., job autonomy), but at the same time have to endure certain demands associated with FWDs (i.e., interruptions). How the experience of these demands and resources differ per individual is also central in this research. It is likely that different personalities relate differently to FWDs and thus relate differently to demands and resources associated with FWDs. Consequently, FWDs can result in different levels of well-being for each individual.

Extensive research has been done on the relationship between personality and various job outcomes (e.g., Barrick & Mount, 1993; Barrick, Stewart, Neubert, & Mount, 1998), but very little research is done on personality in the context of FWDs (Demerouti et al., 2014). Extraverts, for example, perform better at their jobs when having a high degree of autonomy (Barrick & Mount, 1993). Seeing that autonomy over certain work processes is a key aspect of FWDs, it could be that extraverts not only perform better when using FWDs, but also feel better because of it. For the reason that different personalities relate differently to job

outcomes, the following research question is posited: What is the relationship between FWDs

(6)

5 The academic relevance of this research lays in the two key theoretical contributions it has in filling the existing gap in research about FWDs and its influence on an employee’s well-being. First of all, the application of the job demands-resources model in the context of FWDs, expands our knowledge about the relationship between FWDs and well-being. Through the use of the job demands-resources model we gain insight into which job aspects are related to FWDs and which of those aspects are beneficial or possibly disadvantageous for the well-being of employees. Secondly, this research reveals how demands and resources gained from FWDs could depend on personality. With this knowledge, it might be possible to explain certain contradicting findings that exist in research on FWDs. Namely, because it is possible that in these contradicting studies the respondents under research are significantly different, personality wise, from one study to another.

Flexible Work Designs, the Job Demands-Resources Framework and Well-Being

Flexible Work Designs

In the mid-1970’s, during the energy crisis, as a response to the high costs of travelling long distances by car, employees were given the opportunity to access their computers from home through remote terminals over telephone lines (Hesse & Grantham, 1991). We know this particular form of flexible work as telecommuting. A lot has changed and in contradiction to the seventies, nowadays, there are more communication technologies that support and enhance the flexibility people have in where and when they work. The number of people who can decide when and where they work is still growing. In 2009, nineteen percent of the entire working population in the Netherlands were teleworking (Statistics Netherlands, 2009). In 2013 the amount of teleworkers went up to 25 percent (Statistics Netherlands, 2013). There has also been an enormous growth in the Netherlands in ICT use in organizations. In 2009, 61 percent of the employees in the Netherlands use a computer at work with access to internet.

(7)

6 This is double the amount as measured in 2002 (Statistics Netherlands, 2011). Also more than seventy percent of the organizations use some form of intranet, which is expected to increase even more over the years due to, among other things, the introduction of cloud storage possibilities (Statistics Netherlands, 2011).

With the probability in mind that organizations will use communication technology more and more for their work processes, organizations will also be more capable to facilitate employees with FWDs. In many studies on FWDs, the main focus lies on the flexibility of workplace and time of work, while underestimating the importance and role of

communication technology (Demerouti et al., 2014; Ten Brummelhuis et al., 2012). Because of the high dependency of communication technology, in this research a FWD will be defined as a form of working where employees have autonomy over where and when they work, whilst communication with others is facilitated by communication technology (Ten Brummelhuis et al., 2012).

Some examples of advantages of FWDs are that FWDs can reduce the level of absenteeism in an organization (De Menezes & Kelliher, 2011), it could create balance between an employee’s work and family life (Valcour, 2007), and it could result in more job satisfaction (Baltes, Briggs, Huff, Wright, & Neuman, 1999; Kelliher & Anderson, 2010). Although findings on the latter example is not consistent in the literature (De Menezes & Kelliher, 2011). On the other hand, FWDs can also create intensification of work (Kelliher & Anderson, 2010) and stress from interruptions due to communication technology use (Fonner & Roloff, 2012). These are just a few examples of findings in research on FWDs. What the effects of FWDs are on an employee’s well-being is still not fully clear due to contradicting findings (De Menezes & Kelliher, 2011). With the use of the job demands-resources (JD-R) framework (Demerouti et al., 2001), the influence of FWDs on the well-being of employees will be delineated further on.

(8)

7 The Job Demands-Resources Framework

As described earlier on, the job demands-resources framework categorizes working conditions into two broad categories, namely job demands and job resources (Demerouti et al., 2001; Demerouti et al, 2014). Job demands are any aspect of the job that requires

sustained physical and/or psychological effort. These demands only have a negative impact on an person’s well-being when meeting those demands require a lot of effort, of which the person’s is unable to recuperate fully (Demerouti et al., 2001). Job resources are aspects of the job that help with achieving work goals and reduce strain from job demands. Besides that, job resources can also stimulate personal growth, learning and development (Demerouti et al. 2001; Demerouti et al., 2014).

Demerouti et al. (2001) showed that job demands are positively related to exhaustion and that resources are negatively related to disengagement from work (Demerouti et al., 2001). Exhaustion and disengagement from work are considered to be the core dimensions of burnout (Green, Walkey, & Taylor, 1991). Demerouti et al. (2001) also showed that a high amount of job demands not necessarily result in disengagement and that a lack of job

resources not necessarily result in exhaustion. But when a person has limited resources and a high amount of demands, this can create both disengagement and exhaustion. When both of these states are present it represents the burnout syndrome (Demerouti et al., 2001). In this study however, the focus will not be on burnout and its components, but on other psychical and socio-psychological levels of well-being. There will also be a focus on demands and resources that especially apply to FWDs. As shown earlier, there are several paradoxes investigated that are associated with FWDs. The results of these investigations and various other studies will be used to show which aspects of FWDs can be considered as demands and which as resources, and how these demands and resources influence the well-being of

(9)

8 Different Paradoxes and Employees’ Well-Being

As said before, research shows inconsistent findings on the relationship between flexible work and an employee’s well-being (Allen et al., 2013; De Menezes & Kelliher, 2011; Golden, 2009). De Menezes and Kelliher (2011) attributed these mixed findings to the different conceptualizations in research of FWDs. In some studies the focus was on single forms of FWDs (e.g., either workplace flexibility or work time flexibility), while in others no differentiation was made between FWDs and other family-friendly policies. Another reason for the mixed findings are methodological issues like large variations in sample size and homogeneity in sample composition, combined with a lack of consistent reporting of sample information (De Menezes & Kelliher, 2011). Lastly, the mixed findings could be attributed to certain conditions or processes that give us a clearer insight into the relationship between FWDs and well-being (De Menezes & Kelliher, 2011). These conditions and processes will be explored with the paradoxes described in research on FWDs.

The autonomy-control paradox (Putnam et al., 2013), the autonomy paradox

(Mazmanian et al., 2013) and the cycle of responsiveness (Perlow, 2012) will be discussed. The cycle of responsiveness of Perlow (2012) can be viewed as an essential part of both paradoxes. The cycle of responsiveness entails that a person that is using technology to facilitate work on different locations and time, will be perceived as more available by co-workers, will be contacted more by them, and because of that a person will stay connected more to monitor possible unpredictable work developments (Perlow, 2012). The premise of this is that within an organization a culture will be created of being connected all the time. Perlow (2012) revealed in her study that employees spend twenty to 25 hours a week monitoring their work, while not actually working. She also found that employees with a more unpredictable workweek, tended to be more online. Also, as a result of being more online an employee’s work time seemed to become more unpredictable (Perlow, 2012).

(10)

9 A similar process is described in the autonomy-control paradox and the autonomy paradox (Mazmanian et al., 2013; Putnam et al, 2013). Herein an employee gains autonomy by being able to work anywhere and anytime, but loses some autonomy and flexibility because of the increased expectancy of being available and being responsive (Mazmanian et al., 2013; Putnam et al, 2013). This consequently fades the boundary between when work ends and private life starts (Towers, Duxbury, Higgins, & Thomas, 2006), and resulting in an employee working everywhere and all the time (Mazmanian et al., 2013). This premise is in line with the findings of Kelliher and Anderson (2010). They found that employees with flexible work intensify the amount of work they do. For the current study, the intensification of work and unpredictable work developments are viewed as FWD demands. It is found that the intensification of work can induce stress (Macky & Boxall, 2008) and thus lesser a person’s well-being, which is also expected in the case of unpredictability in work.

The premise of the (self-perpetuating) paradox of technology use, is that the use of technology can result in the organization and disorganization of work (Rennecker & Godwin, 2005). Because of technology employees have very easy and fast access to their co-workers. This can result in a decrease in work delays and therefore increase the possibility for

employees to organize their work. Ten Brummelhuis et al. (2012) found a similar relationship, namely a person who works flexible will communicate more effective and efficient with co-workers. She posited that new media technology (e.g., smartphones and email) facilitates efficient time use and coordination with colleagues. Also, it might be that the use of electronic communication, in comparison to face-to-face meetings, happens in a more structured and selective manner. Lastly, Ten Brummelhuis et al. (2012) argue that with electronic

communication, co-workers have less informal conversations than when meeting in person and because email is asynchronous, a person has more time to reflect on received emails and how to formulate their responses. When applied to the paradox of technology use, it seems

(11)

10 likely that effective and efficient communication contributes to the organization of work. As described earlier, technology use can also create a perception with co-workers of increased availability and responsiveness. Consequently, this can result in more interruptions during work (Fonner & Roloff, 2012), thus disorganization. The experience of interruptions with FWDs will be further discussed later on. Effective and efficient communication will be viewed in this study as a FWD resource, because FWDs were found to be negatively related to daily exhaustion and positively related to daily engagement, due to increased effective and efficient communication (Ten Brummelhuis et al., 2012).

In the connectivity paradox, Leonardi et al. (2010) propose that information and communication technologies diminish the perception of distance to others when teleworking, but simultaneously enhance expectations of constant connectivity for employees. This creates a paradox where the potential benefits of teleworking is negated by the same technologies that made the flexible work arrangement possible (Leonardi et al., 2010). Fonner and Roloff (2012) investigated this paradox and looked at an employee’s identification with the

organization and the stress from interruptions that employees might experience as a result of being constantly connected. They found that employees identify themselves less with an organization, because of an increase in stress from interruptions, due to their use of communication technology. Surprisingly, Fonner and Roloff (2012) did not find any significant support that the use of communication media contributes to the perception of social presence and consequently with organizational identification. Because of these findings, interruption is taken under consideration in this research as a FWD demand that is expected to mediate the well-being of employees, when working flexible.

Gajendran and Harrison (2007) have researched the so-called telecommuting paradox. They found in their meta-analysis that because of telecommuting, employees perceive to have more job autonomy and a lower amount of work-family conflict. It was expected that working

(12)

11 on an alternative location could weaken an employee’s social presence in an organization and could therefore diminish one’s interpersonal bonds with co-workers and supervisors, which could hinder possible career advancements (Gajendran & Harrison, 2007). In general, they didn’t find support for damaging effects to workplace relationships or career prospects.

Although, being away from the office for more than 2,5 days per week, did harm relationships with co-workers. Contrary to their expectations, the quality of relationships with supervisors seemed to increase when employees telecommute (Gajendran & Harrison, 2007). The researchers conclude that the telecommuting paradox did not materialize in their study and that only the positive pathway (i.e., increased job autonomy, less work-life conflict) was fully supported (Gajendran & Harrison, 2007). Seeing that Gajendran and Harrison (2007) mainly found support for the positive consequences of telecommuting, job autonomy and work-life balance will be viewed in this study as FWD resources. In relation to an employee’s well-being, job autonomy is found to reduce feelings of stress (Thompson & Prottas, 2006) and balance between work and private life is found to create an overall better assessment with employees of their quality of life (Greenhaus, Collins, & Shaw, 2003).

From the previously described investigations of the paradoxes, certain resources and demands associated with FWDs surfaced. FWDs are expected to positively relate to an employee’s well-being, due to job autonomy, work-life balance, and effective and efficient communication. Interruptions, work intensification, and unpredictability are expected to mediate a negative relationship between FWDs and employees’ well-being. From this the following hypothesis is formulated:

Hypothesis 1: FWDs, operationalized as flexibility over work time (a), workplace (b), and supported by communication technology (c), are positively related to employees’ well-being, through the FWD resources (job autonomy, work-life balance, effective

(13)

12

and efficient communication) and negatively related to employees’ well-being through FWD demands (interruptions, work intensification, unpredictability).

Flexible Work Design Demands and Resources: The Moderating Effect of Personality

The Five Factor Model of Personality

The Five Factor Model of Personality (FFM; McCrae & Costa, 1987) is used in

different research fields and is the most used model to research personality (e.g., Barrick & Mount, 1993; Barrick, Mount, & Judge, 2001). As the name entails, the FFM differentiates between five abstract dimensions of personality, namely extraversion, agreeableness, conscientiousness, neuroticism, and openness. The five dimensions represent the most

abstract levels of personality and each dimension consists of six more specific facets (McCrae & Costa, 1987; Judge, Rodell, Klinger, Simon, & Crawford, 2013). Ones (2005) commented that there is consistent evidence for the hierarchical organization of personality.

Personality is researched extensively in the context of work (e.g., Barrick & Mount, 1993; Barrick et al., 2001), but little research has been done in relation to FWDs (Demerouti et al., 2014). As an example of personality research in an organizational context, Barrick and Mount (1993) showed that managers with a high degree of extraversion or conscientiousness or a low degree of agreeableness performed better in high autonomy jobs. Soane and Chmiel (2005) focused their research on how different personalities relate to preferred risk taking in various areas, work being one of them. They showed, for example, that extraverts prefer taking risks in work-related decisions (Soane & Chmiel, 2005). Also the findings of Leonardi et al. (2010) in their investigation of the connectivity paradox, can give some insight. One of their findings is that teleworkers use different strategies to distance themselves from the office, either by disconnecting or by showing that they are available while being engaged in

(14)

13 other activities. Although these strategies will not be further discussed, it does show that employees differ in the way they use technology. This could be a result of their different personalities, which emphasizes the legitimacy to explore personality in the context of FWDs. In the next paragraphs, the expected influence of the five personality dimensions on the relationship between FWDs and FWD resources and demands will be discussed per personality dimension.

Extraversion

In numerous studies, social isolation is named as a possible disadvantage of workplace flexibility (e.g., Cooper & Kurland, 2002; Golden, 2006). It seems likely that feelings of social isolation could especially occur for extraverts. Extraverted individuals are described as being sociable, assertive, talkative and active (Barrick & Mount, 1991). The broad personality dimension exists of six more specific facets, namely warmth, gregariousness, assertiveness, activity, excitement seeking and positive emotions (Judge, Rodell, Klinger, Simon, & Crawford, 2013). A possible important facet of extraversion in the context of FWDs and social isolation is gregariousness. Individuals that score high on this facet are highly sociable, prefer the company of others, and have the general idea of “the more the merrier” (Costa & McCrae, 1992). Seeing that social isolation could reduce a person’s well-being (House, 2001), this facet could be especially influential for how extraverts experience working at an

alternative location.

For example, Haines, St-Onge, and Archambault (2002) found that telecommuters with a high need for affiliation reported lower work-life quality and telecommuting satisfaction. O’Neill, Hambley, Greidanus, MacDonnel, and Kline (2009) studied the relationship between sociability and the performance and satisfaction of teleworkers in comparison to non-teleworkers. O’Neill et al. (2009) found that sociability was negatively related to the performance of teleworkers, but no difference was found between the job

(15)

14 satisfaction of teleworkers and non-teleworkers. Seeing that both sociability and need for affiliation closely resemble the facet gregariousness (Judge et al., 2013), extraverts might experience a lower work-life quality and lower performance. However, from these studies it is not clear how a FWD influences the job satisfaction of extraverts.

Some insight can be gained with the findings of Clark, Karau, and Michalisin (2012). They investigated the relationship between extraversion and attitudes towards telecommuting amongst business students, and expected a negative relationship due to feelings of social isolation. Contrary to their expectations, Clark et al. (2012) found no relationship between extraversion and attitudes towards telecommuting. According to Clark et al. (2012),

extraverted people might compensate the lack of social interaction with colleagues by social interaction in their private life, which is easier to arrange when having a flexible workplace. That extraverts are inclined to experience more positive effects of FWDs (e.g., social

interaction in private life), instead of negative effects of FWDs (e.g., social isolation from the organization), could be attributed to the positive emotions that extraverts are inclined to have (Judge et al., 2013).

In a study of Judge, Bono, and Locke (2000) the researchers describe that individuals prone to experience positive emotions respond favorably to situations designed to induce positive affect (Judge et al., 2000). Judge, Locke, Durham, and Kluger (1998) describe this notion of favorable responses in an organizational context and note that extraverted

individuals have more positive core self-evaluations and might seek out information in their work environment that would lead to positive conclusions about their work (based on self-verification theory; Swann, Stein-Seroussi, & Giesler, 1992). In the context of FWDs, it is expected that extraverts are more aware of the resources available, because they actively seek out information that would lead to positive conclusions about their flexible work. Therefore it is also expected that highly extraverted employees experience more FWD resources than less

(16)

15 extraverted employees, because less extraverted employees are less inclined to seek out

information due to having less positive core self-evaluations.

This process of extraverts actively seeking out information to confirm positive self-core evaluations also translates to demanding aspects of FWDs. Individuals with positive dispositions rate characteristics of their task or job (e.g., skill variety, task significance) as more enriched than people who are less positively disposed (Judge et al, 2000). With enriched job characteristics the researchers mean that a job is perceived as more challenging (Judge et al, 2000). The reason that extraverts are prone to more challenging jobs is that individuals with more positive core self-evaluations are not only more capable, but also receive more intrinsic rewards when overcoming the challenges they face, in comparison to individuals who have less positive core self-evaluations (Judge et al, 2000). In respect to FWD demands, it is not expected that extraverts experience more demands, but rather less. This is based on how extraverts are able to deal with demanding aspects of their job. Because they are more capable of coping with FWD demands, these demands are likely to be less demanding and therefore less prevalent for highly extraverted employees, in comparison to less extraverted employees.

Another possibly important note with the facet positive emotions is generated from the broad-and-build theory of Fredrickson (2003). This theory suggests that positive emotions broadens people’s modes of thinking and action, which over time builds their enduring

personal and social resources (Fredrickson, 2003). These resources, just as with job resources, act like reserves which help individuals to cope with negative experiences (Tugade &

Frederickson, 2004).

With in mind the workings of positive emotions and positive core self-evaluations, it is expected that highly extraverted employees experience more FWD resources than less

(17)

16 extraverted employees. Also, extraverts are expected to experience less FWD demands, in comparison to employees who are less extraverted. Based on this, the following hypotheses are formulated:

Hypothesis 2: The positive relationship between FWDs, operationalized as flexibility over work time (a), workplace (b), and supported by communication technology (c), and FWD resources (job autonomy, work-life balance, effective and efficient

communication), will be moderated by extraversion, in such a way that extraverted employees experience more FWD resources than less extraverted employees.

Hypothesis 3: The positive relationship between FWDs, operationalized as flexibility over work time (a), workplace (b), and supported by communication technology (c), and FWD demands (interruptions, work intensification, unpredictability), will be moderated by extraversion, in such a way that extraverted employees experience less FWD demands than less extraverted employees.

Agreeableness

Highly agreeable individuals are characterizes as cooperative, trustworthy, amicable, helpful, and honest (McCrae & Costa, 1991). Barrick and Mount (1993) found that managers who were less agreeable performed better than managers who were very agreeable, when having an high amount of job autonomy. One of the explanations that the researchers gave for this finding, is that managers high in agreeableness are more compliant to rules and are more cooperative, and therefore perform better when there is a lot of structure and less ambiguity in how to achieve job objectives (Barrick & Mount, 1993). Because employees low on

agreeableness perform better when having a high amount of autonomy, and employees high on agreeableness might prefer structure over autonomy, it could be that employees low on agreeableness also have more autonomy over their jobs in comparison to employees high on

(18)

17 agreeableness. Therefore, it is expected that employees low in agreeableness experience more job autonomy with FWDs.

Seemingly contradicting, work teams with high mean levels of individual

agreeableness receive high ratings for team performance, social cohesion and low ratings for team conflict (Barrick et al., 1998). Although this seems to contradict the study mentioned earlier where individuals low in agreeableness performed better (Barrick & Mount, 1993), it is not necessarily contradicting. The focus of Barrick et al. (1998) was on the team and their cohesive performance, while the focus of Barrick and Mount (1993) was on the performance of single individuals, namely managers. Also, when looking at the facets of agreeableness (i.e., trust, straightforwardness, altruism, compliance, modesty, and tender-mindedness), it seems plausible that especially people high on agreeableness work well with others. This could be especially beneficial with FWDs, because it is expected that individuals who score high on agreeableness have the tendency to think from someone else’s perspective and accommodate the other. It seems likely, that this tendency of highly agreeable employees to accommodate the other, will translate to the way they communicate with co-workers. It is also likely that an employee with a FWD will have an increase in mediated communication.

Therefore, it is expected that employees high on agreeableness, as a result of their tendency to accommodate others, will have more effective and efficient communication with FWDs, in comparison to employees who are low on agreeableness.

The altruistic focus of highly agreeable employees can also be beneficial for their work-life balance. Wayne, Musisca, and Fleeson (2004) argued that the altruistic and cooperative characteristics of agreeable individuals may reduce the amount of interpersonal tensions and increase the amount of support at work and at home. Although Wayne et al. (2004) didn’t find a negative relationship between agreeableness and family-to-work conflict, they did find a negative relationship between agreeableness and work-to-family conflict

(19)

18 (Wayne et al., 2004). With the notion that highly agreeable employees have less conflict in their family life due to work, in comparison to less agreeable employees, it is expected that in the context of FWDs highly agreeable employees experience more work-life balance.

In relation to FWD demands, it is plausible that because of the altruistic and

cooperative nature of agreeable individuals (Costa & McCrae, 1992), more interruptions from co-workers will occur when working on an alternative location. It might be that co-workers view a highly agreeable colleague as more receptive and available for communication and therefore contact them more in comparison to less agreeable individuals. Although it could be that agreeable individuals don’t necessarily see these interruptions as disruptive, it is expected highly agreeable employees experience more interruptions with FWDs in comparison to less agreeable employees.

With the same reasoning in mind, it might be that the availability and receptiveness for communication with co-workers, when working on alternative locations, could result in co-workers asking more help from employees who are more agreeable. This could lead to an intensification of work for highly agreeable employees. Especially because agreeable people are prone to accommodate others, more so than themselves (Costa & McCrae, 1992). It is plausible that highly agreeable employees are more open for communication, are more available and responsive, and thus monitor their communication technology more for unpredictable work.

To sum up, with FWDs agreeable employees are expected to communicate more effective and efficiently, have a better balance between work and life, experience more work intensification, have more unpredictability at work and receive more interruptions, than less agreeable employees. However, highly agreeable employees are expected to experience less job autonomy. Based on the aforementioned, the hypotheses are formulated as followed:

(20)

19

communication), will be moderated by agreeableness, in such a way that agreeable employees experience less job autonomy, more work-life balance, and more effective and efficient communication, in comparison to less agreeable employees.

Hypothesis 5: The positive relationship between FWDs, operationalized as flexibility over work time (a), workplace (b), and supported by communication technology (c), and FWD demands (interruptions, work intensification, unpredictability), will be moderated by agreeableness, in such a way that agreeable employees experience more FWD demands than less agreeable employees.

Conscientiousness

Conscientious individuals are described as careful, thorough, responsible, organized, self-disciplined, scrupulous, hard working, purposeful, efficient, and precise (Barrick & Mount, 1991). Barrick and Mount (1991) posit in their meta-analyses that conscientiousness is the best predictor of job performance. In line with this, conscientiousness was found to relate positively with job performance, especially when there was a high amount of job autonomy (Barrick & Mount, 1993). According to Barrack and Mount (1993), the better performance is due to the responsible, dependable, persistent and achievement orientated tendencies of conscientious people. These traits are especially useful in those settings where there is a low amount of pressure from the environment a person works in, which can be achieved when having a high amount of autonomy over a job (Barrick & Mount, 1993). Seeing that with FWDs a conscientious employee has a lot of control over the amount of pressure from the environment he or she experiences, in this case also physically, it is

(21)

20 expected that conscientious employees thrive in such a setting. And especially because of their achievement orientated tendencies, it is plausible that conscientiousness employees tend to make more decisions related to their job on their own, when working flexible, because back and forth communication with supervisors would only delay them to achieve work related goals. Therefore, especially for highly conscientious employees, a higher amount job autonomy is expected when working flexible.

Two facets of conscientiousness seem important for effective and efficient communication, namely deliberation and order. Deliberation entails the ability to think carefully before acting and order entails the tendency to be neat, tidy and well organized (Costa & McCrae, 1992). These two facets could contribute to effective and efficient

communication with co-workers, seeing that conscientious employees might be organized in their communication messages and they think carefully before sending it. Thus, when working with FWDs, conscientious employees are expected to communicate more effective and efficient than less conscientious employees.

In respect to work-life balance, Wayne et al. (2004) found a negative relationship between conscientiousness and work-to-family conflict. According to the researchers, careful planning, effective organization, and efficient time management allow conscientious

employees to achieve more in the time that is available. Consequently, this enhances the possibility for an conscientious employee to manage incompatible time pressures, which results in less work-to-home conflicts (Wayne et al., 2004). As described earlier, the traits of conscientiousness are especially useful when having more autonomy (Barrick & Mount, 1993). Thus is expected, in the context of FWDs, that highly conscientious employees experience a better work-life balance in comparison to less conscientious employees.

(22)

21 The self-discipline facet of conscientiousness describes a person’s ability to begin and carry out tasks despite boredom or other distractions (Costa & McCrae, 1992). People who are very disciplined are also described as persistent (Costa, McCrae, & Dye, 1991). This facet, in combination with the facet achievement striving (Costa & McCrae, 1992), might be

important antecedents for the findings of Andreassen, Hetland, and Pallesen (2010). Andreassen et al. (2010) found that conscientiousness was positively related to three

components of workaholism, namely work involvement, drive, and enjoyment of work. This suggests that highly conscientious individuals are more likely to be workaholics and,

therefore, are likely to experience intensification of work as a result of flexibility in workplace and time. On that note, it also seems likely that conscientious employees are more willing and open to take on more tasks because of their drive to succeed. For that reason, it could be that conscientious individuals are more open and available to receive unpredictable work

developments than individuals who are less conscientious. Therefore, highly conscientious employees are expected to experience more work intensification and more unpredictable work developments, in comparison to less conscientious employees.

As described earlier, a conscientious person has the ability to begin and carry out tasks despite boredom or other distractions (Costa & McCrae, 1992). This in combination with being effective in organizing, could result in conscientious employees being more aware of when and how they use their communication technologies. Because of this, it is expected that conscientious employees experience less interruptions from co-workers than less

conscientious employees, due to their conscious deliberation on when to use and when not use their communication technologies.

(23)

22

communication), will be moderated by conscientiousness, in such a way that conscientious employees experience more FWD resources than less conscientious employees.

Hypothesis 7: The positive relationship between FWDs, operationalized as flexibility over work time (a), workplace (b), and supported by communication technology (c), and FWD demands (interruptions, work intensification, unpredictability), will be moderated by conscientiousness, in such a way that conscientious employees

experience less interruptions, more work intensification, and more unpredictability in comparison to less agreeable employees.

Neuroticism

Neurotic people are described as being anxious, easily frustrated, worried, depressed (Costa & McCrae, 1992), insecure, fearful, and emotionally instable (Goldberg, 1990). The facets of neuroticism are anxiety, angry hostility, depression, self-consciousness,

impulsiveness, and vulnerability (Judge et al., 2013).

Clark et al. (2012) found that neurotic individuals, in comparison to emotionally stable individuals, had more favorable attitudes towards telecommuting, preferred telecommuting to conventional work, and perceived telecommuting as presenting less challenges. According to Clark et al. (2012), the reason behind this could be that neurotic workers prefer working off-site to avoid awkward or unpleasant social interactions with co-workers. Working at home could allow neurotic individuals to engage in non-normative behavior without fear of reprisals

(24)

23 from others, and could provide shelter from interpersonal anxieties that might be experienced in the traditional, face-to-face office environment (Clark et al., 2012).

However, when a neurotic person has a positive perception on telecommuting it doesn’t necessarily mean that a neurotic person would thrive under conditions of a FWD. Some perspective can be gained from the study of Judge et al. (2000). As described earlier for positive emotions of extraverts, the opposite can be said for negative emotions of neuroticism. Neurotic individuals are prone to experience negative emotions and negative core

self-evaluations (Judge et al., 2000). In the context of work, neurotic people might seek out and categorize information in their work environment that might attend to negative aspects of their work (Judge et al., 1998). With this notion it seems likely that neurotic workers concentrate on aspects of FWDs that are demanding instead of concentrating on aspects that can be viewed as resources.

Although, neuroticism and negative affectivity are strongly related to work-life conflict (Allen, Johnson, Saboe, Cho, Dumani, & Evans, 2012), and in general FWDs could help accommodate balance between work and life, the negative dispositions of neurotic employees makes it so that they concentrate more on the demands of the FWDs and underutilize, or not utilize, resources to negate these demands.

Based on the negative emotions and focus of neurotic individuals, the following hypotheses are formulated:

communication), will be moderated by neuroticism, in such a way that neurotic employees experience less FWD resources than less neurotic employees.

(25)

24

Hypothesis 9: The positive relationship between FWDs, operationalized as flexibility over work time (a), workplace (b), and supported by communication technology (c), and FWD demands (interruptions, work intensification, unpredictability), will be moderated by neuroticism, in such a way that neurotic employees experience more FWD demands than less neurotic employees.

Openness

Individuals who score high on the dimension openness are depicted as having an active imagination, aesthetic sensitivity, intellectual curiosity, and a preference for variety to the routine (Costa & McCrae, 1992). From the description, especially the preference for variety to the routine, it seems likely that people who are open would prefer FWDs. However, in the study of Clark et al. (2012), no relationship was found between openness and overall telecommuting preferences. Their explanation for this finding is that open individuals might be receptive to various work approaches whether they are experienced in a traditional or telecommuting environment (Clark et al., 2012).

In the study of Barrick and Mount (1993), where they examine the moderating effect of autonomy on the relationship between personality and performance, they exclude openness. Their reason is that this personality dimension yields weak and inconsistent findings for management jobs. Also Bruck and Allen (2003) exclude openness from their study about personality traits and work-life conflict. They posit that the characteristics of openness lack any potential to correlate with work-life conflict. Bruck and Allen (2003) included openness in their investigation only for exploratory purposes, but did not find any relationships.

In relation to the inconsistent findings between openness and job performance, Griffin and Hesketh (2004) posit a possible explanation. They examined the openness construct and suggest that the direction of one’s openness could be either external, where there is a focus on

(26)

25 the environment, or internal, where there is a focus on one’s inner thoughts and feelings (Griffin & Hesketh, 2004). According to the researchers, external openness consists of the facets fantasy, feelings and aesthetics, and internal openness consists of the facets actions, values and ideas. Their proposition was that the two sub dimensions might correlate

differently with job performance. However, in their study Griffin and Hesketh (2004) found only low correlations and most correlations were not significant.

The lack of explanatory power of openness in an organizational context, either when it is related to perception on telecommuting, job performance or work-life conflict, gives some basis to exclude openness from this study. However, in the context of FWDs being newly introduced in an organization, people who are open might adapt better to the new FWD than people who are less open, because of their preference for variety of the routine.

Based on the lack of findings and explanatory power of openness, it is not fully clear how and if openness has a moderating influence on the relationship between FWDs and the FWD demands and resources. However, the dimension is incorporated in the measurement for exploratory purposes, with the following research question: What is the role of openness in

the relationship between FWDs and FWD resources and demands?

Method Sample and Procedure

In total 626 respondents filled in a Dutch web-based questionnaire. When respondents were asked if they had children or not, there was an open answer alternative, in which the answers given by respondents were not usable. Therefore, five cases were excluded from the measurement, leaving 621 respondents in the sample. Half of the sample was female (50.2 percent), and age varied from 18 to 67 years with an average of 45.5 years (SD = 11.33). On average respondents worked 36.4 hours a week (SD = 7.74), had 22.9 years of working

(27)

26 experience (SD = 11.99), and were employed at their current organisation for thirteen years (SD = 10.77). In the sample, 19.6 percent (123 respondents) of the respondents worked in healthcare, 14.1 percent (88 respondents) had a government job, 9.3 percent (58 respondents) worked in the industrial sector, 8.5 percent (53 respondents) worked in business services, and the remaining 48.5 percent worked in various other sectors like the financial sector, hotel and catering industry, education, and transportation. Almost a quarter of the sample, 24.8 percent (155 respondents), indicated to have a management position. More than half of the

respondents (58.9 percent) indicated having at least one child of which 72.2 percent (267 respondents) at least have one child living at home, and 65.7 percent (411 respondents) of all respondents live together with a partner.

Measures

All scales that were developed were reliable and the scores of the reliability are shown in the diagonal of Table 1. The lowest Cronbach’s Alpha measured was for the scale of work

intensification (α = .84). The highest Cronbach’s Alpha measured was for the scale of flexibility over work time (α = .96).

Flexible work designs

The New Ways of Working scale of Ten Brummelhuis (2011) was used to assess the measures of autonomy over work time and workplace. In the original scale, there are three items that measure autonomy over work time and three for autonomy over workplace. Respondents were able to indicate on a five-point Likert scale if they agreed with a statement. A score of one indicated strong disagreement and a score of five indicated strong agreement. An example of an item measuring flexibility over work time is: ‘I decide myself when my workday starts’. The average score on the scale was 2.1 (SD = 1.23). An example of an item measuring

flexibility over workplace is: ‘I decide myself where I work’. The average score on the scale

(28)

27 Communication technology use

Communication technology was assessed with the use of five items, of which each item relates to a different communication technology. These technologies were telephone, videoconferencing (e.g., Skype), instant messaging, e-mail and groupware (e.g., Dropbox, Cloud-services). Respondents were asked to rate on a five-point Likert scale to what extent (1

= very little, 5 = very much) they used the previously mentioned technologies to perform their

jobs accordingly. The items were added to form an index scale of communication technology use. The minimum possible score was therefore five and the maximum was 25. On average the respondents scored 12.0 (SD = 4.19).

Job resources

There are three kinds of FWD resources used in this study: job autonomy, work-life balance, and effective and efficient communication. Job autonomy was measured with three items from the Work Design Questionnaire (WDQ; Morgeson & Humprey, 2006). An example of an item is: ‘The job gives me a chance to use my personal initiative or judgement in carrying out the work’. Respondent indicated on a five-point Likert scale if they agreed with a statement. The average score on the scale was 3.5 (SD = 0.97). Work-life balance was used with a five-item scale developed by Valcour (2007). Respondents were asked to indicate on five-point Likert scale, from one (very dissatisfied) to five (very satisfied), how satisfied they were with the statements proposed. An example of an item is: ‘The way you divide your time between work and personal or family life’. The average score on the scale was 3.6 (SD = 0.83). Effective and efficient communication was also measured with five items, on which respondents indicated to either agree or disagree on a five-point Likert scale. An example of an item is: ‘The way I communicate with my colleagues is usually effective’. A principal component analysis (PCA) showed that the five items formed a single uni-dimensional scale. There was only one component that had an eigenvalue above 1 (eigenvalue = 3.49). The scree

(29)

28 test confirms this seeing that there was a clear point of inflexion after the first component. The variance explained by the component was 69.8 percent. The average score on the scale was 3.8 (SD = 0.71).

Job Demands

The three FWD demands are interruptions, work intensification and unpredictability of work. Interruptions was measured with five items and respondents could indicate on a five-point Likert scale to what extent they agreed with a statement. An example of an items is: ‘Sometimes I find it difficult to concentrate on my work because of all the interruptions’. A PCA was conducted to verify if the five items form a one-dimensional scale. The PCA indicates that the five items exists of one component. The eigenvalue was 3.64 and the component explains 72.9 percent of the variance. Also the scree test clearly indicated the presence of only one component. There was an average score of 2.8 (SD = 0.98). Six items were used to asses work intensification and respondents were able to indicate on a five-point Likert scale to what extent they agreed with the statements. The six items came from a scale developed by Macky and Boxall (2008). An example of an item is: ‘I never seem to have enough time to do all my work’. The average score on the scale was 2.8 (SD = 0.96).

Unpredictability of work was assessed with the use of six developed items on which

respondents were able to indicate on a five-point Likert scale to what extent they agreed with the statements. An examples of an item is: ‘The nature of my work sometimes results in unpredictable working pressure’. To verify if the items form a one-dimensional scale, a PCA was conducted. The PCA indicated that the six items formed a one-dimensional scale with an eigenvalue of 3.95 and explained variance of 65.9 percent. Also the scree test indicated the presence of one component. On the scale the average score was 3.3 (SD = 0.92).

(30)

29 Personality was measured with the 240 items of the NEO-PI-R (Dutch version) of Costa and McCrae (1992). Each personality dimension was assessed with 48 items. An example of an item of extraversion, agreeableness, conscientiousness, neuroticism, and openness are respectively: ‘I like most people I meet’, ‘I try to be courteous with everyone I meet’, ‘I keep my stuff neat and clean’, ‘I am easily scared of something’ and ‘I have a lively imagination’. On each statement respondents could indicate to what extent they agreed with a statement. For each personality dimensions the reversed items were recoded and the total 48 items were added to form index scales. The minimum score on each scale was 48, with a maximum of 240. For the scale of extraversion the mean was 150.6 (SD = 20.72), for agreeableness was the mean 168.3 (SD = 18.17), for conscientiousness 166.8 (SD = 18.18), for neuroticism the average was 128,6 (SD = 23.74), and for openness 151.3 (SD = 16,97).

Well-being

Well-being was assessed with the use of two validated scales. First the general health of respondents was measured using the General Health Questionnaire of twelve items (GHQ-12). The GHQ-12 was developed by Goldberg and Williams (1988). An examples of an item is: ‘Were you recently able to concentrate well on what you do?’. Respondents were able to answer on a five-point Likert scale whether they experienced the proposed state of well-being a lot less than usual (one) or a lot more than usual (five). Six of the twelve items were

reversed and recoded. There was an average score on the scale of 1.6 (SD = 0.68). The higher the score of a respondent on the scale, the higher their well-being was. The second measure of well-being was with the Somatic Complaints subscale of the Dutch version of the Symptom Checklist (SCL-90) (Arrindell & Ettema, 1986). The subscale of Somatic complaints exists of eight items and respondents were able to indicate on a five-point Likert scale if they

experienced the somatic complaint asked with either totally not (one) or a lot (five). An example of an item is: ‘To what extent were you hindered this last week by painful muscles?’.

(31)

30 The scale had a mean score of 1.5 (SD = 0.68). The higher a respondent scored, the more somatic complaints the respondent had, and the worse their physical well-being was. Analysis

The mediation analyses for the hypotheses 1a, b and c, were conducted with the use of Hayes’s (2013) PROCESS macro (Model 4) for SPSS. The six mediators were tested by calculating bias-corrected confidence intervals of 95 percent, with a thousand bootstrap samples. The PROCESS macro only allows one independent variable per analysis and there were three independent variables under study. Therefore each analysis consisted of one independent variable with the remaining two used as co-variates to be able to asses the full model measurement. The macro also allows only one dependent variable, for which there is no workaround.

The moderation analyses for the hypotheses 2 to 9, were also conducted with the use of SPSS. For the moderations, multiple regression analyses were conducted, with

standardarized variables and standardarized interaction terms. For the interpretation of the two-way interaction effects the plot procedure of Dawson (2014) was followed. There were ninety regression analyses conducted to asses each interaction effect seperatly for each FWD component and each job demand and job resource variable.

Results

Descriptive statistics

In Table 1 the standard deviations, the means and the correlations of all variable are presented. In total there were eight demographic variables which were used as control

variables in every analysis conducted. These are age, gender, co-habiting, children, size of the organisation, years in service for the organisation, the working hours per week and managerial position. All of these demographic variables correlated, although mostly slighty, with the dependent variables or with some of the FWD resources and demands, and therefore were

(32)

31 included in the analyses. Strong positive correlations exist between communication

technology use and FWD time (r = .40, p<0.05) and FWD place (r = .38, p<0.05), which is expected seeing that flexible workers are likely to be dependent on these communication technologies for performing their job. Also, neuroticism has strong negative correlations with conscientiousness (r = -.61, p<0.05) and extraversion (r = -.50, p<0.05).

Insert Table 1 about here

As can been seen in Table 1, the job demands (interruptions, work intensification, unpredictability) have strong positive correlations with eachother (ranging from: r = 55,

p<0.05, to: r = 65, p<0.05). The job resources (job autonomy, work-life balance, effective and

efficient communication) also correlate with eachother, only these are weak to moderate positive relationships (ranging from: r = 27, p<0.05, to: r = 37, p<0.05).

Mediation results

In Table 2 and Table 3 all direct and indirect relationships are presented for each FWD component. There were no indirect effects found for flexibility over time on employees’ general health, through any of the FWD demands or resources (see Table 2 and Figure 1). There were also no indirect effects found for flexibility over time on employees’ somatic complaints (see Table 3 and Figure 1). Therefore mediation hypothesis 1a is not supported.

The mediation hypothesis 1b is partially supported. Flexibility over workplace yields a significant positive indirect effect on employees’ general health, through job autonomy (b = .006, BC 95% CI [.001, .016]), and work-life balance (b = .018, BC 95% CI [.005, .036]).

(33)

32 Also, flexibility over workplace had a significant negative indirect effect on employees’ somatic complaints, also through job autonomy (b = -.008, BC 95% CI [-.022, -.001]), and work-life balance (b = -.013, BC 95% CI [-.028, -.003]). No significant indirect effects were found through effective and efficient communication or the FWD demands (see Table 2 and 3, and Figure 1). Thus, overall, flexibility over workplace had a positive indirect effect on the well-being of an employee, through job autonomy and work-life balance.

Communication technology use had a significant positive indirect effect on employees’ general health, through job autonomy (b = .003, BC 95% CI [.001, 006]). Communication technology use had a significant negative indirect effect on employees’ somatic complaints, also through job autonomy (b = -.004, BC 95% CI [-.008, -.000]), and effective and efficient communication (b = -.004, BC 95% CI [-.009, -.000]). A significant negative indirect effect of communication technology use on employees’ general health was found, through interruptions (b = -.006, BC 95% CI [-.011, -.003]). Lastly, a significant positive indirect effect of communication technology use on employees’ somatic complaints was found, through work intensification (b = .005, BC 95% CI [.002, .010]). Thus, overall, communication technology use had a positive indirect effect on an employee’s well-being, through job autonomy and effective and efficient communication, and a negative indirect effect on an employee’s well-being through interruptions and work intensification. No

indirect effects were found through, work-life balance and unpredictability (see Table 2 and 3, and Figure 1). Therefore, hypothesis 1c is partially supported.

(34)

33 Insert Figure 1 about here

In relation to the direct effect, as are shown in Figure 1, flexibility in time was not directly related to any of the FWD demands or resources. Flexibility over place was only positively directly related to job autonomy and work-life balance. Communication technology use had a positive relationship with job autonomy and effective and efficient communication, interruptions, work intensification and unpredictability (also see Table 2 and 3). These relationships will be further discussed in the discussion section.

Moderation results

The results of the regression analysis of the moderating effects of extraversion on the relationship between FWDs and FWD resources are shown in Table 4.

The positive relationship between flexibility over work time and the FWD resources is only moderated by extraversion in the case of effective and efficient communication (t (13, 607) = -2.314, p<0.05). Figure 2 shows the directions of the relationship. In general, highly extraverted employees had more effective and efficient communication when having a high or low amount of flexibility over work time, in comparison to less extraverted employees. But, less extraverted employees increased their effective and efficient communication when having a high amount of flexibility over work time, in comparison to their effective and efficient communication when having a low amount of flexibility over work time. For highly

(35)

34 extraverted employees, their effective and efficient communication decreased as their

flexibility over work time increased. Therefore, hypothesis 2a is not supported.

Insert Figures 2 about here

Hypothesis 2b is partially supported. Only the relationship between flexibility over workplace and work-life balance was moderated by extraversion (t (13, 607) = 2.329, p<0.05). Figure 3 depicts this moderation effect. In general, highly extraverted employees experienced more work-life balance when having flexibility over workplace, in comparison to less extraverted employees. When the amount of flexibility over workplace increases, highly extraverted employees experience an increase in their work-life balance. For employees low on extraversion there doesn’t seem to be a difference in their work-life balance when working with a high or low amount of flexibility over workplace.

Insert Figure 3 about here

Hypothesis 2c was not supported. There were no moderation effects found for

extraversion on the relationship between communication technology use and any of the FWD resources (see Table 4).

In Table 5 the moderation effects of extraversion on the relationships between the FWD components and the three FWD demands are shown. Hypothesis 3a and 3b are both not supported. Extraversion did not moderate the relationship between flexibility over work time or workplace and any of the FWD demands (see Table 5).

(36)

35 Insert Table 5 about here

Hypothesis 3c is fully supported. Extraversion moderates the relationships between communication technology use and interruptions (t (13, 607) = -3.352, p<0.01), work intensification (t (13, 607) = -2.765, p<0.01), and unpredictability (t (13, 607) = -1.981,

p<0.05), all in such a way that highly extraverted employees experienced less of the FWD

demands, in comparison to less extraverted employees. These moderation effects are shown in Figure 4, 5, and 6.

Insert Figure 4, 5 and 6 about here

When communication technology use increased, less extraverted employees

experienced a bigger increase in all of the FWD demands, in comparison to the increase of FWD demands for more extraverted employees.

In Table 6 the regression results are shown for the moderation effects of agreeableness on the relationship between the FWD components and the three FWD resources.

In regard to the personality trait agreeableness, hypotheses 4a was not supported, because the only moderation effect found of agreeableness on the relationship between flexibility over work time and effective and efficient communication was not in the expected direction (t (13, 607) = -2.010, p<0.05). The direction of this moderation effect is shown in Figure 7.

(37)

As the amount of flexibility over work time increased, the amount of effective and efficient communication that more agreeable employees had decreased. For less agreeable employees the amount of effective and efficient communication experienced, didn’t differ when having a low or high amount of flexibility over work time.

Hypotheses 4b and 4c were both not supported. There were no moderation effects found of agreeableness on the relationship between flexibility over workplace, or

communication technology use, and the FWD resources.

Table 7 shows the regression results of the moderation effects of agreeableness, on the relationships between the FWD components and the three FWD demands. Hypotheses 5a and b are not supported. There were no moderation effects found of agreeableness on either the relationship between flexibility over work time and the FWD demands, or on the relationship between flexibility over workplace and the FWD demands.

Hypothesis 5c is also not supported, because there was one significant moderation effect found of agreeableness on the relationship between communication technology use and work intensification (t (13, 607) = -3.224, p<0.01), and the direction was not as expected. In Figure 8 the direction of this moderation effect is shown. Less agreeable employees

experienced more work intensification when using communication technology, in comparison to highly agreeable employees. For highly agreeable employees there was no difference in work intensification when either using a low or high amount of communication technology.

(38)

For hypotheses 6a, b and c, the moderating effect of conscientiousness was found in the relationship between flexibility over work time and effective and efficient communication (t (13, 607) = -2.617, p<0.01), between flexibility over workplace and work-life balance (t (13, 607) = 2.406, p<0.05), and between communication technology use and effective and efficient communication (t (13, 607) =-2.031, p<0.01). However, for the remaining parts of the hypotheses, no moderation effects were found (see Table 8). In Figures 9, 10, and 11, the directions of the moderation effects of conscientiousness found, are shown.

Insert Figures 9 and 10 about here

It was expected that conscientiousness would moderate the relationships between FWDs and effective and efficient communication, in such a way that highly conscientious employees would experience more effective and efficient communication, than less

conscientious employees. However, both with flexibility over work time and communication technology use, the directions were not as expected. For both highly conscientious employees and less conscientious employees, the amount of effective and efficient communication decreased, as the amount of flexibility over work time increased, but this decrease was more prevalent for highly conscientious employees. In respect to communication technology use, both highly conscientious employees and less conscientious employees, experienced more

(39)

38 effective and efficient communication, but this increase in effective and efficient

communication was more apparent for less conscientious employees. Based on the aforementioned, hypotheses 6a and 6c are not supported.

Insert Figure 11 about here

In respect to hypothesis 6b, the moderation effect of conscientiousness on the relationship between flexibility over workplace and work-life balance, was in the expected direction (see Figure 11). Highly conscientious employees experience more work-life balance when having flexibility over workplace, in comparison to less conscientious employees. For less conscientious employees no difference was found between a high or low amount of flexibility over workplace, and their experience of work-life balance. Therefore, hypothesis 6b is partially supported.

In Table 9 the moderation effects of conscientiousness is shown for the three FWD demands. Hypotheses 7a and b are not supported. There were no moderation effects found of conscientiousness on the relationships between flexibility over work time or workplace and any of the FWD demands.

Hypothesis 7c is partially supported. Conscientiousness moderated the relationship between communication technology use and interruptions (t (13, 607) = -2.069, p<0.05), in such a way that highly conscientious employees experience less interruptions, in comparison to less conscientious employees. Both highly conscientious and less conscientious employees