Is home-based teleworking really effective? : a study about the possible influence of the amount and age of children, gender and extraversion on the relationship between home-based teleworking, job performance and organ

(1)

MASTER THESIS BUSINESS STUDIES Leadership and Management

By Renée Rijnsdorp – 6061214 Supervisor: Dr.C. Ashton James

Date: 27th June 2014

IS HOME-BASED TELEWORKING REALLY

EFFECTIVE?

A study about the possible influence of the amount and age of children,

gender and extraversion on the relationship between home-based

(2)

2

Abstract

This study focuses on the relationship between home-based teleworking and job performance,

and between home-based teleworking and organizational commitment. Previous research

about home-based teleworking is mixed. Some studies argue that home-based teleworking has

a lot of benefits like improved work-family balance and an improved performance

(McNamara et al., 2013, Caillier, 2012). While other studies list disadvantages like

work-family conflict and social isolation (Morganson et al., 2010; Gajendran & Harrison, 2007;

Panaccio & Vandenberghe, 2009).

This study focuses on the effect of home-based teleworking on job performance and

organizational commitment and investigates which factors can influence these relationships.

In this study three moderators were tested: the age and amount of children at home, gender

and the degree of extraversion.

A moderator analysis was conducted in a sample of 180 home-based teleworkers from

different companies. The amount and the age of the children influenced the relationship

between home-based teleworking and job performance; this relationship was significant when

there were three children or more at home. Also gender had a significant negative influence

on the relationship between home-based teleworking and job performance, in particular that

men show lower productivity when they are working from home. Extraversion didn’t have an

influence on the relationship between home-based teleworking and job performance. The

moderators tested in this research didn’t influence the relationship between home-based

teleworking and organizational commitment.

Keywords: Home-based teleworking, job performance, organizational commitment, gender, children, extraversion

(3)

3

1. Introduction

No more suits, no distracting colleagues, no more travelling but working in your pyjamas at

home whenever you want. Governments are encouraging telecommuting by implementing

legislation (Gajendran & Harrison, 2007; FNV, 2012). Working from home seems like a

normal thing to do these days and has even become a trend that is likely to continue in the

upcoming years (NRC, 2012). In 2013 already one third of the employees in the Netherlands

is working from home for an average of six hours per week (CBS, 2013). Home-based

teleworking or telecommuting has three components namely the location of the workplace is

autonomous from the location of the employer, information technology is used such as

telephones and computers, and there is a link to the organization (Baruch, 2000).1 Working

from home offers employees the flexibility to manage their own time and thus spend more

time with their families. Since more and more people start working from home it is important

that the implications are clear. Research has shown that home-based teleworking has positive

as well as negative implications.

Home-based teleworking has benefits for both the employers and the employees. It

enhances productivity, reduces costs as well as absenteeism, while employees manage their

own time and thus have a more balanced work-family life (Golden, 2006). When people work

from home, organizations save on office and travelling expenses of employees and as a result

it is cost effective (Hill et al., 2008; Baruch, 2000; Ward & Shabba, 2001). At the same time

there are also benefits for the employees since it allows them more flexibility in their life and

thus it is likely that their well-being will increase. Both men and women try to balance the

needs of work and family life and often experience job-related stress (King et al., 2009;

Vandello et al., 2013). Work flexibility gives employees the possibility to regulate the

1

Home-based teleworking is a form of work flexibility. Flexible work arrangements are options that allow employees to work at different times (flexitime) and at different places (flexplace) (Allen et al., 2013). In this research the emphasis will lie on home-based teleworking since this is the most common used form of work flexibility and most research that has been done sees home as the remote location where people work (Baruch, 2000; Gajendran & Harrison, 2012).

(6)

6

demands between work and family, which has a positive impact on the work-family balance

of the employees and their performance (McNamara et al., 2013, Caillier, 2012). It thereby

reduces work-related stress because employees don’t have to travel to work anymore

(Gajendran & Harrison, 2007). Home-based teleworking can thus have a positive impact on

the well-being of the employees, well-being is characterized by emotional components like

pleasant affects and cognitive components like job satisfaction (Panaccio & Vandenberghe,

2009). Employees working from home experience feelings of freedom, increased autonomy,

reduced distractions and are more committed to their organization (Gajendran & Harrison,

2007; Golden, 2006). Teleworking can thus provide a win-win situation for the employer as

well as its employees (Wheatley, 2012).

But despite the positive findings about home-based teleworking it can have

disadvantages as well and therefore has to be analysed with caution. Working from home can

lead to an increased work family conflict, social isolation, invisibility to supervisors and thus

career stagnation (Wheatley, 2012). Since family and work life occurs at the same time and

place, it is likely for one domain to intrude the other leading to more work-family conflict

(Gajendran & Harrison, 2007). This problem occurs more for women since they have children

to take care for and tend to work from home more often than men (Wheatley, 2012; Hilbrecht

et al., 2008). When the work-family life gets blurred it is harder for employees to separate the

activities between home and work. People working from home thereby miss the social context

of work since the face-to-face communication with their colleagues is reduced, which can

lead to social isolation (Morganson et al. 2010, Gajendran & Harrison, 2007). Social isolation

can have a negative effect on employee well-being since it can cause stress and work-family

conflict (Panaccio & Vandenberghe, 2009).

Research has shown that employees are more productive when they are working from

(7)

7

Harrison, 2007; Golden et al., 2008). Both the quality and the quantity of the work that is

done when working from home is better (Baruch, 2000). This can be explained by the fact

that there is less distraction at home, people can adjust their work to their environment and

save travel time, as a result of which they are able to work more effectively (Baruch, 2000;

Gajendran & Harrison, 2007). Employees who work from home are also more motivated,

loyal and engaged to their organizations (Hill et al., 2008). Home-based teleworking increases

employee well-being because employees feel more satisfied since they have more autonomy

and it responds better to their needs (Gajendran & Harrison, 2007, Wheatley, 2012). People

who work from home are therefore more committed to their organization (Hunton & Norman,

2010; Golden, 2006). But since home-based teleworking can also have negative influences

like an increased work-family conflict and social isolation, it is likely that this has a negative

effect on employee well-being and it thus decreases the job performance and organizational

commitment. Since the existing literature about home-based teleworking is contradictory and

ambiguous it is therefore important that more research is conducted about home-based

teleworking. This research will further investigate the conditions under which home-based

teleworking may not be beneficial. This will be analyzed by conducting a questionnaire in a

sample of home-based teleworkers. The research question of this thesis is:

What factors influence the relationship between home-based teleworking and job performance and between home-based teleworking and organizational commitment?

This study focuses on two concepts that are related to home-based teleworking. The first

concept is job performance. Job performance is a multi-dimensional construct and in this

research it will be measured through two dimensions namely proficiency and the work quality

(8)

8

fulfilled and the employees meet the expectations, also known as task performance (Griffin et

al., 2007). Work quality is more subjective since it measures how much effort employees put

in their work and what the quality of the outcomes is (Kuvaas & Dysvik, 2009). The second

concept, organizational commitment is related to job performance (Winkler, 2012; Hunton &

Norman, 2010). It usually results in a supportive organization and a willingness to promote

the organization’s well-being (Meyer & Allen, 2001). The emphasis in this research will be

on affective commitment since this has a strong positive relationship with job performance

and is considered as the most important component of organizational commitment (Hunton &

Norman, 2010). Affective commitment is defined as the strength of the identification and

involvement of the employee in a particular organization and is positively related to employee

well-being (Panaccio & Vandenberghe, 2009). Employees with a high affective commitment

stay with an organization because they choose to do so (Hunton & Norman, 2010).

This study aims to give a more thorough insight into the relationship between home-based

teleworking and job performance and between home-based teleworking and organizational

commitment. This will provide more knowledge about which conditions influence these

relationships. In this research the influence of gender, children and extraversion will be

investigated. As Golden (2006) suggested, more research needs to be done about the specific

aspects of individuals like environmental circumstances (i.e. the number of children at home)

and personality differences when people are working from home. However, there is no

empirical evidence yet if these factors influence the relationships. This information is also

important for organizations since home-based teleworking has an effect on job performance

and organizational commitment. This research will clarify if other factors influence these

relationships, as a result of which organizations can determine whether it is beneficial for

(9)

9

This study starts with an overview of the factors that influence the relationship between

home-based teleworking and between job performance and organizational commitment.

Subsequently three factors that could possibly influence the relationship will be discussed, in

this research the factors that will be further investigated are the amount and age of the

children at home, gender and the degree of extraversion. Based on the literature review a

theoretical model has been developed and empirically tested in order to come to an answer of

(10)

10

2. Theoretical foundation

In this section the theoretical background of this study will be outlined, after which the

hypotheses will be formulated. In order to examine the effect of home-based teleworking on

job performance and on organizational commitment, it is important to know which factors

have an influence on the effectiveness of home-based teleworking. Secondly, it will be

analyzed whether the factors gender, children at home and extraversion have an influence on

the relationship between home-based teleworking and job performance, and between the

relationship between home-based teleworking and organizational commitment.

2.1 The effectiveness of home-based teleworking

For teleworking to be effective various factors have to be present according to Baruch (2000)

and Kowalski & Swanson (2005). Examples of these factors are the nature of the job and the

fit with technology, a supportive organization, the interface with home and work and the

individual fit. The factors are divided into four categories namely support, communication,

trust and individual factors.

2.1.1 Supporting home-based teleworking

For home-based teleworking to be successful, support must come from all levels within the

organization (Kowalski & Swanson, 2005). A supportive culture is a culture that supports and

encourages home-based teleworking and causes support throughout the organization

(Cegarra-Leiva et al. 2012). A lack of support can have negative outcomes like social

isolation (Ward & Shabba, 2001) and negative career consequences (Cegarra-Leiva et al.,

2012). Managers are often resistant to home-based teleworking since they lack trust or still

have a traditional attitude that doesn’t support working from home (Kowalski & Swanson,

2005). This can lead to career stagnation of the teleworkers. It is therefore necessary that an

organization develops management approaches that are supportive of home-based

teleworking. The development of HR practices can enhance the autonomy and responsibility

(11)

11

(Sanchez et al., 2007). Organizing trainings for the managers where they learn the benefits of

home-based teleworking helps the management to create positive attitudes towards it

(Kowalski & Swanson, 2005). Supporting home-based teleworking leads to higher employee

satisfaction, higher job performance, improved work-family balance, increased organizational

commitment and thus increases organizational outcomes like productivity and quality

(Cegarra-Leiva et al., 2012; Kowalski & Swanson, 2005).

Due to recent advances in technology like high-speed internet access, instant

messaging and web conferencing it is possible to work from remote locations and thus

enables working from home (Hunton & Norman, 2010). Working from home is not possible

for jobs where physical presence is necessary, for example when contact with clients is

necessary or for the co-ordination of the productivity of the organization (Baruch, 2000). It is

critical for the success of home-based teleworking that the organization gives technological

support and provides the right equipment to the employees (Kowalski & Swanson, 2005). A

training for the teleworkers on how to use the equipment, manage their time and establish the

boundaries between their work and family life can help employees to improve their

performance while working from home. Especially support to establishing the boundaries

between their work and family life is important for the employees since this can reduce stress

(Kowalski & Swanson, 2005).

The performance of a teleworker can be affected by the way colleagues and family

members look at their work status (Baruch, 2000). Family members often provide the

teleworker help in their work by answering the phone, taking messages or giving practical

support (Sullivan & Lewis, 2001). A supportive family is thus of great importance since they

understand and respect the work, leading to a better work-family balance, more satisfaction

(12)

12

What also plays an important role in the effectiveness of home-based teleworking is

the communication in an organization, which will be discussed next.

2.1.2 Communication

When people work from home the communication between the manager and the employees

becomes more complicated (Kowalski & Swanson, 2005). It is therefore important to develop

new communication skills such that the teleworkers don’t feel isolated. It is crucial that the

teleworker still feels a part of the organization and knows what can be expected.

Communication can play a huge role in determining whether home-based teleworking is

effective.

Organizations should develop policies to enhance the communication between the

employees. A formal teleworking policy should be developed to communicate issues

regarding safety, privacy, health and security to the employees (Kowalski & Swanson, 2005).

Besides the formal communication that informs the teleworkers about work related matters,

informal communication is necessary to prevent social isolation. Informal communication

allows employees to develop relationships and ensures that they feel a part of the

organization. Managers can arrange meetings, conference calls or social events to stimulate

the informal communication among home-based teleworkers. Intranet can also play an

important role by making the teleworkers feel a part of the organization, by having access to

updates and announcements. Home-based teleworkers who regularly visit the office

experience less social isolation (Bailey & Kurland, 2002). As a result, another way to prevent

social isolation is by mixing home-based working with office-based working (Ward &

Shabba, 2001). Employees will then still experience the social interaction of the traditional

work experience.

Social isolation is one of the biggest disadvantages of working from home and it

(13)

13

Swanson, 2005). Communication leads to trust and the other way around, the role of trust will

be discussed in the next section

2.1.3 Trust

Another crucial factor for successful home-based teleworking is trust at all levels of the

organization, as a result of which the organizational culture should be based on trust

(Kowalski & Swanson, 2005). Organizations are more likely to offer home-based teleworking

to employees they trust (Sanchez et al., 2007; Bailey & Kurland, 2002). Some managers use

control mechanisms by which they can monitor and control the communication and the work

of the teleworkers (Baruch, 2000). But the management also has to trust their employees that

they are doing their job well when they are working from home, despite the fact that they

can’t directly see them. Managers need to trust their employees that they are doing what is

expected from them (Kowalski & Swanson, 2005).

Also the teleworkers have to trust their manager that they will be treated equally and

that the interpersonal treatment is fair, also known as interactional justice (Saunders &

Thornhill, 2002). The way the teleworkers are treated influences their trust, and if there is a

lack of trust it is likely that home-based teleworking will not be successful. Trust can lead to

better performance, job satisfaction and lower job related stress. If the organizational culture

is based on trust, the teleworkers will be assessed on their performance and the completion of

their assignments, instead of attendance (Kowalski & Swanson, 2005; Baruch, 2000). A

culture that is based on trust is the result of commitment and identification with the shared

values of an organization and needs to be built through employee empowerment. When

employees throughout the organization trust each other it is likely that home-based

teleworking will be more effective.

Support, communication and trust are interrelated (Kowalski & Swanson, 2005).

(14)

14

organization leads to trust. Besides these three factors also the individual qualities and

circumstances play a role, this will be discussed next.

2.1.4 Individual factors

Since no person is the same, home-based teleworking has a different impact on people. The

attitude, values, norms, qualities, needs and the work-home interface of a teleworker can

influence the performance (Baruch, 2000; Ward & Shabba, 2001). Personality and skills

differences can affect the acceptability of working from home. People with high needs for

autonomy like working from home, while people with high needs for affiliations may

experience it as a disadvantage. Whether employees are intrinsically motivated to carry out

their job and their self discipline determine if home-based teleworking is effective (Ward &

Shabba, 2001). For example, teleworkers with more self-efficacy tend to perform better

(Sanchez et al., 2007). Since teleworkers in general are social employees they adapt quickly to

working from home as soon as they are familiar with the disadvantages and know what to

expect (Ward & Shabba, 2001). These disadvantages and how to deal with them can be

learned in trainings (Kowalski & Swanson, 2005).

To stay motivated, it is important for employees that they enjoy their work and feel

challenged by it (Ward & Shabba, 2001). Motivated employees want to achieve results and

are committed to work for this. Employees see performance as a reward for their efforts.

These rewards can be divided in intrinsic rewards like self-esteem, and extrinsic rewards like

status efforts (Ward & Shabba, 2001). What individuals perceive to be a fair reward for their

work differs per person and has an effect on their job satisfaction and thus on their

performance and organizational commitment. It is therefore important that teleworkers should

be encouraged to use their skills when they are working from home, in order to perform

(15)

15

The effectiveness of working from home also depends on the meaning and the impact

the work has on the individual. In some situations home-based teleworking is a good idea but

it has to fit both the person and the home situation (Baruch, 2000). The home-work interface

covers the quality of the family relations and the facilities that are available when working

from home. (Kowalski & Swanson, 2005). To effectively work from home a separate room is

needed, so the size of the house of the teleworker also plays a role (Baruch, 2000). In case

teleworkers are working in a non-suitable work location, their motivation and performance

will decrease. To create an appropriate and stimulating environment for home-based

teleworkers, organizations should give employees a funding so they can buy suitable furniture

and equipment (Ward & Shabba, 2001).

It can thus be concluded that a lot of individual factors play a role in determining if

home-based teleworking is effective. It has to fit the personality and the home situation. In the

next section three individual factors will be highlighted namely children at home, gender and

extraversion.

2.2 Potential factors influencing job performance and organizational commitment

A lot of factors influence the effectiveness of home-based teleworking. It is not only

important that colleagues and management throughout the organization are supportive, but

also that family, friends and colleagues of the teleworker are supportive (Kowalski &

Swanson, 2005). This means that the organization has to trust their employees that they will

perform well when they are working from home. To make sure people in the whole

organization trust each other it is important that the communication is clear about home-based

teleworking. So the organization plays a huge role in determining whether home-based

teleworking will be effective for their employees, the culture needs to be build on trust

(Baruch, 2000; Kowalski & Swanson, 2005). Besides that, also some individual factors have

(16)

16

attributes that influence which people will be effective when teleworking still needs to be

determined. In this part three potential factors will be analyzed which can possibly influence

the relationship between based teleworking and job performance and between

home-based teleworking and organizational commitment namely the amount and the age of the

children at home, gender and extraversion of the employees.

2.2.1 Children at home

Teleworking can have an impact on the work-family balance, which influences employee

well-being. Well-being can be defined as a positive feeling of how people experience their

life. Research has shown that well-being is related to job performance (Wright & Huang,

2012). Work-family conflict is a predictor of a negative employee well-being (Grant-Vallone

& Donaldson, 2001). Work-family conflict has negative effects like turnover and stress and is

negatively related to job, life, marital and family satisfaction as well as to organizational

commitment (Allen et al., 2013). Overall it can be concluded that home-based teleworking is

negatively related to work-family conflict since employees are better able to regulate their

work and family demands (Gajendran & Harrison, 2007). But sometimes the boundaries

between work and family life get blurred because teleworkers work in overtime or are not

able to manage the personal freedom of teleworking (Sullivan & Lewis, 2001). Since

work-family conflict has an effect on employee well-being and thus on the performance and

commitment of the employees, it is important to know what influences this relationship.

For some employees, working from home is the only way to overcome the high costs

of childcare (Sullivan & Lewis, 2001). But in this case the children are at home while people

are working. Having children at home often leads to fragmented working hours since the

teleworkers organize their workday around the children and family life (Hilbrecht et al., 2008;

Sullivan & Lewis, 2001). It seems logical that children create more work-family conflict for

(17)

17

working from home (Baruch, 2000). Parents with young children are having a hard time

combining the work and family responsibilities when working from home (Currie & Eveline,

2011; Bailey & Kurland, 2002). Carlson (1999) found that the number of children at home has

a strong effect on the work-family conflict. Madsen (2003) didn’t find a significant effect

when employees have children when they are teleworking, but she didn’t take into account

whether the children are present at home during the teleworking hours or not.

Therefore in this research the amount of children at home will be taken into account

since this appears to have an impact on the work-family conflict. It can be expected that

younger children will lead to less job performance. This leads to the following hypothesis:

Hypothesis 1a: The relationship between home-based teleworking and job performance is moderated by the amount and the age of children at home, such that having more children and younger children at home will attenuate the relationship between home-based teleworking and job performance.

Since having children has an influence on the work-family conflict, it is likely that it will have

an influence on the organizational commitment of home-based teleworkers. Women with

children tend to be less involved in their job, which can lead to a lower organizational

commitment (Korabik & Rosin, 1995). Since people with children at home tend to experience

more work-family conflict they will be less satisfied and less involved in the job. It is thus

likely that they will be less committed to their organization. Therefore this leads to the

(18)

18 Hypothesis 1b: The relationship between home-based teleworking and organizational commitment is moderated by the amount and the age of children at home, such that having more children and younger children at home will attenuate the relationship between home-based teleworking and organizational commitment.

2.2.2 Gender

More women are entering the workforce because of an academic degree. For women often

childcare is a reason to work from home (Sullivan & Lewis, 2001). Since women are still

responsible for the biggest part of the childcare and other tasks at home, female employees

will thus face more work-family conflict than men (Ko et al., 2013). It is thus likely that

women will use home-based teleworking as a means to balance the work and family

responsibilities. Women are having a harder time than men to balance the work-family life

since they are doing more housework than men and have not reduced their professional tasks

or responsibilities (Currie & Eveline, 2011). Although the domestic participation of men is

slowly increasing it doesn’t mean that the traditional distribution of domestic work is

changing (Sullivan & Lewis, 2001). Men don’t necessarily participate more in the household

responsibilities when they are working from home.

Gender also has an effect on the way people deal with work flexibility (Ko et al.,

2013). Both men and women have to deal with the competing demands when working from

home. But men and women tend to deal with these demands differently. Men often have a

separate room when they are working from home and in addition their partners often ensure

that they are not distracted from their work (Sullivan & Lewis, 2001). While women usually

work in a shared room and have to look after their children at the same. Men often get more

help from family members when they are working from home than women do. It is thus

harder to manage the work and family demands for women, resulting in a negative impact on

(19)

19

will have a different influence on job performance for men than it has for women. This leads

to the following hypothesis:

Hypothesis 2a: The relationship between home-based teleworking and job performance is moderated by the gender of the employees, such that this relationship will be positive for men and negative for women.

Women often believe that teleworking has a positive effect on their well-being (Hilbrecht et

al., 2008). But since women still do most of the domestic work they experience more

work-family conflict when they are working from home, this can lead to a negative well-being

(Sullivan & Lewis, 2001; Panaccio & Vandenberghe, 2009). Negative well-being can lead to

dissatisfaction, and this can influence the organizational commitment of an employee and lead

to intentions to leave the organization. Because women who work from home experience

more work-family conflict it is suspected that women will be less committed to the

organization when they are working from home. This leads to the following hypothesis:

Hypothesis 2b: The relationship between home-based teleworking and organizational commitment is moderated by the gender of the employees, such that this relationship will be positive for men and negative for women.

2.2.3 Degree of extraversion

Social isolation due to teleworking tends to have a negative impact on the employee’s

well-being and it is thus likely that it will have a negative effect on job performance. Research has

shown that personality influences job performance (Minbashian et al., 2009). More research is

needed to understand how individual differences influence the effectiveness of home-based

(20)

20

People will act differently under teleworking circumstances because of individual

differences (Ward & Shabba, 2001). Being highly introverted might affect the individual

reactions to home-based teleworking (Golden, 2006). This also has an effect on how

employees will react on social isolation that might be experienced when they are working

from home. People can feel socially isolated when they are separated from their colleagues

and have little or no contact during their working hours. In the office the environment is busy

and full of distractions and people, while at home you are alone. According to Ward &

Shabha (2001) a major motivation for people go to work is to experience social contacts.

When people work from home, the social factor goes away which can result in personal

isolation. But this is not always the case, since it depends on the level of social needs of the

individual, which in turn is related to the individual’s personality. One aspect of personality is

the degree of extraversion, being defined as the level of engagement an individual initiates

with the environment. Someone who is an extravert will seek a lot of verbal and non-verbal

contact with others, is sociable, assertive and feels energized by other people (Lang et al.,

2012). In the workplace, extraverts are trusted by their colleagues and experience pleasant and

social interactions with other people (Panaccio & Vandenberghe, 2012). It is likely that

extraverts develop strong social relationships and this will motivate them in their job, leading

to higher job performance. Introverts are more reserved, quiet, shy and can feel drained by

others. Introverts are more likely to work on their own at the office (Lang et al., 2012).

Extraverts work better under arousing circumstances while introverts work better when there

is less arousal (Matthews et al., 1989). Research has shown that introverts can deal better with

social isolation than extraverts do (Tranel, 1962).

Keeping this in mind it is more likely that extraverts will work better at the office

while introverts work better from home where there is less arousal. No studies have been done

(21)

21

previous research it can be expected that extravert people will perform less when they are

working from home. This leads to the following hypotheses:

Hypothesis 3a: The relationship between home-based teleworking and job performance is

moderated by extraversion, such that this relationship is weaker for higher values of extraversion.

Extraverts tend to experience more involvement in the organization and feel reciprocated to

their organization (Erdheim et al., 2006). Extraversion is significantly positive related to

organizational commitment (Panaccio & Vandenberghe, 2012; Erdheim et al., 2006). In

particular to affective commitment because it represents the positive emotional reaction of an

employee to the organization and this is at the core of extraversion (Erdheim et al., 2006). If

people work from home they will have less face-to-face contact with their colleagues and

supervisors. It can therefore be expected that extraverts who work from home will be less

committed to their organization. This leads to the following hypothesis:

Hypothesis 3b: The relationship between home-based teleworking and organizational commitment is moderated by extraversion, such that this relationship is weaker for higher values of extraversion.

(22)

22

3. Research Model

In Figure 1 depicted below are the hypotheses drawn from the theory in a framework.

H1 H2 H3

Figure 1: Theoretical framework Home-based teleworking • Degree of Telework Outcomes • Job Performance • Organizational commitment Degree of extraversion Children at home Gender

(23)

23

4. Methodology

In this section the methodology of this study will be presented. First the research approach

and strategy will be briefly explained, then the data collection technique, followed by the

measurement of the variables and finally the data analysis and conclusions.

4.1 Research approach and strategy

This study is deductive since it clarifies the theory and searches for relationships between

home-based teleworking and job performance, and between home-based teleworking and

organizational commitment (Saunder & Lewis, 2012). The choice for a deductive study is

based on the fact that the purpose of this study is to test theoretical propositions. To better

understand the relationship between home-based teleworking and job performance and

organizational commitment, the conditions that influence this relationship are analyzed

making this study exploratory. Exploratory studies aim to seek new insights, this study tries to

assess home-based teleworking in a new light (Saunder & Lewis, 2012).

In order to test the hypotheses, a quantitative research strategy is used namely, a

survey design in the form of a questionnaire. A survey design is also known as a

cross-sectional design, which is the collection of data on more than one case at a single point in time

(Bryman, 2008). This research strategy makes it able to collect data in a cost effective way

from a large number of people. The data that is collected in the survey will be used to

statistically examine and explain the relationships (Saunder & Lewis, 2012).

4.2 Questionnaire

The type of questionnaire used for this research was self-administered, since people were able

to identify for themselves whether they wanted to complete the survey (Saunders & Lewis,

2012). Self-perception can be a reliable and valid source of information on performance

(Baruch, 2000). But self-evaluations also tend to be skewed, since people have positive

(24)

24

original survey is in English because the content of the questionnaire that are used comes

from literature written in English. Since the survey is distributed on the internet, the survey

had to be available in Dutch and in English for possible international employees. The

questionnaire was translated into Dutch with the help of two native speakers. This was done

in order to reach a large number of respondents. To make sure the survey was correctly

translated to Dutch, it was also translated back into English. This is called ‘the back and

forward method’, which increases the validity of the items (Field, 2009).

The questions in the survey came from other researchers’ questionnaires. All the

questions used in the questionnaire were closed or have a Likert scale ranging from 1

(strongly disagree) to 5 (strongly agree). After the closed questions, some demographic data

was collected like the respondent’s age, gender, tenure, management level, nationality, hours

working and the age and amount of children.

Prior to the distribution of the questionnaire, a pilot test was done to check if all the

questions would likely to be understood (Saunders & Lewis, 2012). Three students

(Bachelor’s and Master’s Degree) and two telecommuters were asked to give comments and

suggestions on the questionnaire.

4.3 Research Sample

The Centraal Bureau Statistiek (CBS) found in 2013 that one third of the employees in the

Netherlands are working from home for an average of 6 hours per week (CBS, 2013). About

7,3 million people are working in the Netherlands, leading to around 2,4 million people who

are working from home. Since this population is very large, it was difficult to locate a

sampling frame. The sample of this research consisted of 180 people who regularly work from

home. In this study, people work from home for at least 6 hours per week. This fits the

research question since it measures the home-based teleworkers. Of these respondents, 54,2%

(25)

25

were Dutch, the other 22% had other nationalities. Their organizational tenure was mostly

(50,7%) between 0 and 5 years. The age of the respondents was between the 21 and 60 years

old. Half of the respondents had children (49%) and the other half did not have children

(51%). Of the respondents 96% worked 0 – 10 hours per week from home and most

respondents (71,5%) worked more than 31 hours per week. The degree of telework was

calculated by the hours of telework divided by the total working hours to calculate the

percentage of telework, as can be seen in Figure 2. It was concluded that 40% of the people

work 0% - 25% of their time from home, whilst the percentage of respondents that works

from home full-time was 14,6%.

Figure 2 The degree of telework in percentages

During a month the participants were recruited with an online survey. The survey was

constructed with the help of the Qualtrics website (Qualtrics, 2014), which made it possible to

distribute the survey on the internet. Respondents were recruited through internet-based social

networks such as forums for teleworkers and LinkedIn, this was done carefully since not all

forums are about working for an organization. So the respondents were selected using

non-40 15,9 19,5 9,8 14,6

Degree of telework

0 - 25% 26 - 50% 51 - 75% 76 - 99% 100%

(26)

26

probability sampling, within this non-probability sampling two methods were used namely

self-selection sampling and snowball sampling. People could choose for themselves whether

they want to fill in the survey online, which can be typified as self-selection sampling. The

respondents were also asked whether they could pass it on to their friends or colleagues who

are also working from home, which can be typified as snowball sampling (Bryman, 2008).

In addition, email messages were sent to the HR department of companies who allow

working from home, explaining why the research would be useful for these companies. A free

research consultancy was offered to the participating companies. Unfortunately, only one

company replied positively and asked their employees to fill in the survey. This can be

explained to the fact that home-based teleworking is a very popular topic at the moment and

companies replied that they get a lot of requests for research on this topic.

4.4 Measurement of variables

In the questionnaire only validated scales of other researchers were used since this increases

the content validity and construct validity. It thereby increases the reliability of the research

and makes it possible to compare the results with other studies which have used the same

scales (Saunders & Lewis, 2012). To make sure that the questionnaires used in this research

reflect the construct that it is measuring, the Cronbach’s Alpha of these scales have to be

above 0.7 (Field, 2009). Since the questionnaire was conducted in two different language

versions, the Cronbach’s Alpha of the Dutch and the English version have to be congruent.

The independent variable that was used was the hours of home-based teleworking. The

moderator variables were gender, age, amount of children at home and the degree of

extraversion. The dependent variables were job performance and job commitment. The

(27)

27

Independent variable

Hours of home-based teleworking: ask reports of an average workweek spent working at home (Golden et al., 2006). The question that was asked is: ‘indicate the proportion of an

average workweek spent teleworking.’ The answers ranged from 0-10 hours, 11-20 hours,

21-30 hours and 31 hours or more. It was also asked how many hours per week they work. The

hours of telework were then divided by the total working hours to calculate the percentage of

telework.

Moderator Variables

Demographics: at the end of the survey the respondents filled in a list with demographics like gender, age, education, years working at the organization, organizational level, amount and

age of children at home and hours working per week. The moderator variables gender and

amount and age of children were obtained with the demographics.

Degree of extraversion (10 items): were measured with a scale of the HEXACO-60 (Ashton & Lee, 2009). The original HEXACO Personality Inventory consists of a

questionnaire of 192 items, since this takes too long to complete a brief HEXACO Inventory

was used. There is a high reliability in using these items and the HEXACO-60 is

recommended when the time is brief for filling out the questionnaire. So the HEXACO-60

scale was adapted with an internal consistency of 0.77 in English and 0.79 in Dutch. An

example item of the Ashton & Lee scale (2009) is ‘On most days I feel cheerful and

optimistic’ and ‘I feel reasonably satisfied with myself overall’. Items were rated on a scale ranging from 1 (strong disagree) to 5 (strongly agree).

(28)

28

Dependent variables

Job performance (21 items). There are many different ways to measure the performance of individuals. In this research job performance was measured with proficiency and the work

quality, consisting of work effort and the work quantity. A 10 item scale of Kuvaas & Dysvik

(2009) was used which measures the work effort and the work quality. The internal

consistency of this scale is 0.9 in Dutch and 0.91 in English. An example of one of the items

of the Kuvaas & Dysvik scale (2009) is. ‘The quality of my work is top-notch’. Work effort

and work quality were rated on a scale ranging from 1 (strongly disagree) to 5 (strongly

disagree). Proficiency was measured using a 7 item scale of Griffin et al. (2007) with an

internal consistency of 0.93 in Dutch and 0.86 in English, the items were rated in terms of

how often in the months these tasks were carried out: never, rarely, regularly, often or very

often. An example of the Griffin et al. scale (2007) is ‘Carried out the core parts of you job

well’.

Organizational commitment (8 items). Organizational commitment was measured with a 8 item scale for affective commitment of Allen & Meyer (in Hunton & Norman, 2010). It is

the most used scale for organizational commitment. The Cronbach’s Alpha is 0.83 for the

Dutch scale and 0.78 for the English scale. Examples of the scale are ‘I would be very happy

to spend the rest of my career with this organization’ and ‘I really feel as if this organization’s problems are my own’. Items were rated on a scale ranging from 1 (strongly agree) to 5 (strongly disagree)

4.6 Control variables

In order to minimize unintended effects that may influence the results, a number of control

variables were introduced. When these variables are held constant, a more adequate

(29)

29

done (Field, 2009). The demographic control variable was age and the job related control

variables were tenure and organizational level.

The first control variable was the age of the respondents since this will likely influence

the dependent variables (Golden, 2006; Ko et al., 2013). Age was measured by asking

respondents to indicate their age in years using 6 categories.

The data was controlled for the tenure, or the amount of years someone works within

the same organization. Research has shown that this influences the job performance and

organizational commitment (Loi et al., 2012; Golden, 2006).

The third control variable was the organizational level of the employee since this

appears to affect the performance of employees. Organizational level was measured by asking

the respondents whether they were top-management, middle-management, low-management

or non-management. The organizational level has an influence on both the dependent

variables (Conway & Briner, 2012). It is likely that people with a higher organizational level

are more committed to the organization and want to perform better (Loi et al., 2012).

4.7 Ethics

In this research, attention was given to ethics in order not to harm people through whom the

research was conducted (Bryman, 2008). Participation in this study was voluntary and

respondents were able to quit the research whenever they felt like it. In addition, the

anonymity of the respondents was guaranteed. In order to make the respondents more

comfortable with the research, a brief motivation was given for this study and the email

(30)

30

5. Results

In this section the results of the research are discussed. First it is explained how the data is

prepared, then the descriptive statistics will be given followed by the correlations. Finally, the

hypotheses can be tested using a moderation analysis.

5.1 Data preparation

To make sure that the questionnaires used in this research reflect the construct that it

measures, the Cronbach’s Alpha of these scales have to be above 0.7 (Field, 2009). Since the

questionnaire was in two different languages, the Cronbach’s Alpha of the Dutch and the

English version have to be congruent. In table 2 the Cronbach’s Alpha of the scales are

shown. All the Cronbach’s Alpha’s are higher than 0.7 and so no action was undertaken.

The variables included in this research were the degree of extraversion, work effort,

work quality, proficiency and organizational commitment. First the variables were calculated,

whereby negative items were recoded. Then the variables were computed by taking the mean

of the items of the variable. To say something about the dependent variable performance, in

the end work effort, work quality and proficiency were computed since they measure

performance. Next, the missing data was coded with 999. The missing data was handled

according to the ‘Pairwise deletion’ meaning that only the cases without missing data in each

pair of variables are being analyzed (Bryman, 2008). Though this has to be done with caution

since the sample size that is related to every coefficient may vary. This can lead to biased

estimates and inconsistent results.

5.2 Descriptive statistics

All the variables except proficiency were measured with a 5-point Likert scale where 1

represents ‘strongly disagree’ and 5 represents ‘strongly agree’. Proficiency was measured

(31)

31

and the standard deviations can be found in Table 2. For extraversion the mean is 3,56, which

means that the respondents saw themselves overall as extravert persons. The mean for the

performance variables (work effort, work quality and proficiency) were also above 3

indicating that respondents rated their performance relatively high. The mean for

organizational commitment was a bit lower namely 3,21 but still above 3 indicating that the

respondents were quite committed to their organization. The standard deviation for all

variables were between 0,64 and 1,69 which means that the variance within the variables is

reasonable.

An analysis was done in order to investigate whether the data is normally distributed

around the centre of all scores, the results of this skewness and kurtosis analysis can be found

in Table 1. The data for proficiency and organizational commitment were normally distributed

since they are close to 0. Extraversion, work effort and work quality have positive values of

kurtosis which indicate a pointy distribution (Field, 2009). The negative skewness for the

same three variables indicated that the distribution is skewed to the left and indicated a

build-up of high scores. The Kolmogrov-Smirnov test was then done to test whether the distribution

as a whole was different from a comparable normal distribution (Field, 2009). This test

showed that the values for all the variables, except for performance, were significant

(p < .001) meaning that the data of these variables were not normally distributed. By making

boxplots of the variables it was determined whether there are outliers in the data. Since

outliers tend to skew the distribution this impact needs to be reduced (Field, 2009). Due to

this reason one respondent has been deleted and excluded from the analysis. However, this

(32)

32

Skewness Kurtosis Kolmogrov-Smirnov

Extraversion -1,523 4,376 0,000 Work Effort -1,244 2,134 0,000 Work Quality -1,400 2,587 0,000 Proficiency -0,557 -0,527 0,007 Commitment -0,623 -0,667 0,000 Performance -0,90 1,28 0,05

Table 1 Test of the distribution of the data and normality

After the normality analyses, multicollinearity between the variables was tested.

Multicollinearity is a strong correlation between two independent variables and should be

avoided (Field, 2009). In this research there was no multicollinearity between the independent

variables. This was confirmed by measuring the variance inflation factors (VIF) which were

all lower than 3 in this research and so there was no problem of multicollinearity.

5.3 Correlations

In Table 2, the correlations, means and standard deviations for all the variables are presented

and show the different significant correlations.

The correlation matrix shows that the degree of telework correlates significantly

negative with the different variables of performance namely effort (r = -0.30, p<0.01), quality

(r = -0.31, p<0.01) and proficiency (r = -0.18, p<0.05). So there is a negative correlation

between the degree of telework and job performance (r = -0.18, p<0.01). Furthermore, from

this matrix it can be observed that the degree of telework correlates significantly negative

with organizational commitment (r = -0.19, p<0.05).

Some of the possible moderators also correlate with the dependent variables. The

amount of children is significantly negative correlated with performance (r = -0.41, p<0.01).

Also the age of the children is negatively correlated with work effort (r = -0.24, p<0.05),

suggesting that the amount of children and the age of the children influences performance.

(33)

33 M SD 1 2 3 4 5 6 7 8 9 10 11 12 Gender 1,54 0,5 Age 3,24 1,3 -0,2* Telew. 48,38 32,78 -0,09 0,01 Tenure 1,92 1,11 -0,18* 0,33** 0,19* Am Ch 2,2 1,59 -0,22* -0,03 0,31** 0,39** Age Ch 2,76 1,6 -0,09 0,26* -0,17 0,27** 0,36** Extrav 3,56 0,64 0,09 0,11 -0,33** -0,11 -0,51** -0,13 (0.78) Effort 3,54 0,87 0,15 0,01 -0,30** -0,28** -0,59** -0,28* 0,59** (0.90) Qual 3,5 0,78 0,17 0,1 -0,31** -0,18* -0,59** -0,24* 0,54** 0,73** (0.90) Prof 4,7 1,69 0,09 0,03 -0,18* -0,15 -0,22** -0,15 0,43** 0,50** 0,58** (0.90) Com 3,21 0,8 0,05 0,08 -0,19* -0,01 -0,22** -0,03 0,43** 0,51** 0,45** 0,30** (0.80) Perf 4,14 0,99 0,13 0,11 -0,18* -0,22** -0,41** -0,81 0,63** 0,72** 0,74** 0,95** 0,43** Manlev 2,54 1,3 0.25** -0.28** -0.25** -0.36** -0.44** -0.25* 0.02 0.14 0.09 -0.11 -0.17 -0.1

Table 2 Means, Standard Deviations and Correlations among Variables

M = Mean, SD = Standard Deviation

*Correlation is significant at the 0.05 level (2-tailed) **Correlation is significant at the 0.01 level (2-tailed)

(r = -0.22, p<0.01). Extraversion is significantly positive correlated with performance

(r = 0.63, p<0.01) and with organizational commitment (r = 0.43, p<0.01).

Of the control variables only tenure correlates significantly negative with job

performance (r = - 0.22, p<0.01). Both management level and age also seem to correlate

with organizational commitment and job performance. Therefore those variables will be taken

into account when conducting regression analysis.

5.4 Testing the hypotheses

To test the relationships between the variables and the hypotheses, a linear regression analysis

and a moderation analysis was done. A moderation tests under which circumstances the effect

exists, to test a moderation the program ‘Process’ was used (Hayes, 2014). A significance

level of p<0.05 was used to confirm or reject a hypothesis.

To test the hypotheses, first a regression was done to test the linear relation between

the dependent and the independent variable. The first relationship was tested with a regression

(34)

34

relationship between home-based teleworking and job performance is significantly positive

(r = 0.37, p<0.05). The control variables are age and tenure and management level. The model

as a whole is significant (F = 3.91, p<0.05). The explained variance in job performance,

dedicated by the degree of telework (β= -0.16, t = -1.98, p<0.05) is 14% (R² = 0.14). Meaning that people perform better when they are working from home.

Then a regression analysis was conducted to test the linear relation between

home-based teleworking and organizational commitment, the same control variables were used.

There is a positive relationship between home-based teleworking and organizational

commitment (r =0.31, p<0.05). The model as a whole seems significant (F = 4,17, p<0.05).

The explained variance in organizational commitment, dedicated by the degree of telework

(β= -0.24, t = -2.53, p<0.01) is 9% (R² = 0.09). Thus home-based teleworking leads to increased organizational commitment. The amount of telework is associated with greater

organizational commitment. Table 3 shows a summary of the results of the regression

analyses. Independent variable Dependent variable Control variables Beta t p R² Degree telework Job performance Age Tenure ManLevel -0.16 -1.98 0.05 0.14 Degree telework Organizational Commitment Age Tenure ManLevel -0.24 -2.53 0.01 0.09

Table 3 Regression results (direct effects)

To test the hypotheses (H1a – H3b), the group of moderators was entered into the regression

analysis with the program ‘Process’. Since this research aims to understand what the

conditions are when an effect operates, a moderation was done. The results are presented in

(35)

35

Independent variable

Dependent variable

Moderator variable Beta t p R²

Degree telework

Job performance Number of children Age of children -0.01 -0.00 -3.85 -1.43 0.00 0.15 0.25 0.12 Degree telework Organizational Commitment Number of children Age of children 0.00 -0.00 -0.02 -0.07 0.98 0.94 0.06 0.03 Degree telework

Job Performance Gender 0.01 2.29 0.02 0.08

Degree telework Organizational Commitment Gender 0.00 0.74 0.46 0.04 Degree telework

Job Performance Extraversion 0.01 1.53 0.13 0.45

Degree telework

Organizational Commitment

Extraversion -0.00 -0.39 0.69 0.19

Table 4 summary of the regression results (moderating effects)

To test H1a: The relationship between home-based teleworking and job performance is

moderated by the amount and the age of children at home, such that having more children and younger children at home will attenuate the relationship between home-based teleworking and job performance. A moderation analysis was done. The interaction model is significantly negative between the degree of teleworking and job performance for the amount

of children (F = 15.17, p<0.00). The amount of children is thus found to be a moderator

(β= -0.01, t = -3.85, p<0.01), this can be seen in Figure 3. Meaning that the more children people have at home, the less they perform when teleworking. The effect is significant for

(36)

36

three children or more, so this means if people have three or more the less the job

performance is when they work at home.

Figure 3 Amount of children as a moderator

However, this interaction is not significant for the age of the children (F = 3,38, p>0.15). And

the age of children is thus not found to be a moderator (β= -0.00, t = -1.43 p>0.05). So the age of the children at home does not moderate the relationship between home-based teleworking

and job performance, but the amount of children does. Therefore hypothesis H1a is partially

supported.

To test H1b: Hypothesis 1b: The relationship between home-based teleworking and

organizational commitment is moderated by the amount and the age of children at home, such that having more children and younger children at home will attenuate the relationship between home-based teleworking and organizational commitment, a moderation analysis was conducted. The interaction model is not significant for the amount of children (F = 2.28,

p>0.98) and it is also not significant for the age of the children (F = 0.62, p>0.94). So the

amount of children is not a moderator (β= 0.00, t = -0.02, p>0.05), and neither is the age of children a moderator (β= -0.00, t = -0.07, p>0.05). Thus the amount and the age of the

2,8 2,9 3 3,1 3,2 3,3 3,4 3,5 Jo b P erf orma nc e Home-based Teleworking Children 0 -3 Children 3 +

(37)

37

children is not moderating the relationship between home-based teleworking and

organizational commitment. Therefore, hypothesis 1b is not supported.

To test H2a: The relationship between home-based teleworking and job performance is

moderated by the gender of the employees, such that this relationship will be positive for men and negative for women, a moderation analysis was conducted. The interaction model is significantly positive for gender (F = 3,99, p<0.02). Gender is thus found to be a moderator

(β= 0.01, t = 2.29, p<0.01), this can be seen in Figure 4. In the interaction model it can be found that this relation is significantly negative for men (β= -0.01, t = -3.09, p<0.01). So this means that the relationship between home-based teleworking and job performance is

moderated by gender, such that this relationship is weaker for men. This means that men tend

to perform less when they work from home. Since this is the opposite of what was expected,

hypothesis 2a is rejected.

Figure 4 Gender as a moderator

To test H2b The relationship between home-based teleworking and organizational

commitment is moderated by the gender of the employees, such that this relationship will be positive for men and negative for women, a moderation analysis was conducted. The

2,8 3 3,2 3,4 3,6 3,8 4 4,2 4,4 4,6 Jo b P erf orma nc e Home-based Teleworking Women Men

(38)

38

interaction model is not significant (F = 1,59, p>0.46). So gender is not to be found a

moderator of the relationship between home-based teleworking and organizational

commitment (β= 0.00, t = 0.74, p>0.05). Hypothesis H2b is thus rejected.

To test H3a The relationship between home-based teleworking and job performance is

moderated by extraversion, such that this relationship is weaker for higher values of extraversion,, a moderation analysis was conducted. The interaction model is not significant (F = 30.02, p>0.45). The analysis thus shows that extraversion is not moderating the

relationship between home-based teleworking and job performance (β= 0.01, t = -1.53, p>0.05). Therefore, H3a is not supported.

To test H3b The relationship between home-based teleworking and organizational

commitment is moderated by extraversion, such that this relationship is weaker for higher values of extraversion, a moderation analysis was conducted. The interaction is not significant between the degree of teleworking and organizational commitment for extraversion (F = 8.58,

p>0.19). Thus extraversion is not moderating the relationship between home-based

(39)

39

6. Discussion

Working from home is a trend these days and more and more organizations start

implementing it, resulting in a significant increase in the number of employees working from

home. Research has shown that home-based teleworking can have positive as well as negative

effects. But it is not clear when working from home is beneficial and when it is not. In this

study the relationship between home-based teleworking and job performance, and the

relationship between home-based teleworking and organizational commitment has been

analysed. More specific, the moderation effect of three individual factors has been examined;

namely the amount and age of children at home, gender and extraversion. Although not all the

results presented in the previous section were significant, some interesting relationships have

been found which are worth investigating in further research.

In this section, the results are discussed in order to be able to draw conclusions. Furthermore,

an answer is given to the research question of this study: What factors influence the

relationship between home-based teleworking and job performance and between home-based teleworking and organizational commitment? Finally, some limitations of this study are given, as well as a number of managerial implications and suggestions for further research.

6.1 Discussion of the results

The effectiveness of home-based teleworking can be influenced by a lot of factors, both

organizational and individual. In this research some of these individual factors have been

investigated to further our knowledge about home-based teleworking.

The presence of children at home simultaneous with home-based teleworking tasks

often leads to fragmented work hours and has an effect on the work-family conflict (Hilbrecht

et al., 2008; Sullivan & Lewis, 2001; Carlson, 1999). Especially young children require more

(40)

40

affecting the effectiveness of working from home (Baruch, 2000; Currie & Eveline, 2011;

Bailey & Kurland, 2002). It was therefore predicted that the amount and the age of the

children has an effect on the performance of employees. This was partially confirmed in this

research. The amount of children did have a significant negative influence on the relationship

between home-based teleworking and performance of teleworkers, if people have more than

three children at home their work performance is reduced. This is also supported by Carlson

(1999), who states that the amount of children affects the work-family conflict. The age of the

children was negatively correlated with work effort, meaning that employees with younger

children put less effort in their work. But the age of children didn’t turn out to be a moderator

of the relationship between home-based teleworking and job performance. So for this research

it can be concluded that the more children the home-based teleworkers have, the less they

perform but that the age of the children at home doesn’t have a notable influence.

Due to the traditional distribution of domestic work, women are still more responsible

for the child care and other tasks at home (Ko et al. 2013; Sullivan & Lewis, 2001). Because

women perform more domestic work than men when working from home, it was predicted

that this has a negative impact on the job performance for female home-based teleworkers.

Surprisingly, in this research the opposite from what was expected emerged namely that this

has a negative impact on the job performance of men. This means that the relationship

between home-based teleworking and job performance is moderated by gender, such that this

relationship is weaker for men. Meaning that men perform less when they are working from

home.

Extraverts tend to work better under arousing circumstances (Matthews et al., 1989).

Extravert people experience pleasant social interaction at the work place and this motivates