What is the impact of late night work-related vs. social
smartphone use on cyberloafing and work-home interference,
and are these effects moderated by self-control?
MSc Business Administration Leadership & Management Master Thesis
Author: Mandy Leenen
UvA ID: 11145846
Thesis Supervisor: Dr. Merlijn Venus
2 Statement of Originality
This document is written by Student Mandy Leenen who declares to take full responsibility
for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources
other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of
3 Abstract
The world has become increasingly interconnected. This constantly being connected can have
various consequences on our daily lives. This study explored the effects of late night
work-related and social smartphone use on work-home interference that same evening and
cyberloafing the next day. The moderating effect of self-control on these possible
relationships was also investigated. 84 participants filled out a one-shot survey and daily
surveys were distributed twice a day during ten working days. Results indicated that
work-related smartphone use after work hours had a positive significant effect on work-home
interference that same evening and that social smartphone use had a marginally positive
significant effect on cyberloafing the next morning. Even though not all expected
relationships were found, the results of the current research extend the already existing
literature on the effects of smartphone use and thereby represents a thorough basis for future
research.
Key words: work-related smartphone use, social smartphone use, work-home interference
4 Table of contents
Abstract ... 3
Table of contents ... 4
List of Tables and Figures ... 5
Introduction ... 6
1. Theoretical Background ... 9
1.1 Smartphone use after work hours ... 9
1.2 Cyberloafing ... 10
1.3 Work-home interference ... 13
1.4 Self-control ... 15
2. Research Method ... 19
2.1 Research Design ... 19
2.2 Sample and Procedure ... 19
2.3 Measures ... 20 2.4 Analysis ... 22 3. Results ... 24 3.1 Correlations ... 24 3.2 Hypotheses Testing ... 25 3.4Additional Analysis ... 28 4. Discussion ... 30 4.1Theoretical Implications ... 31
4.2 Limitations and Future Directions ... 33
4.3 Practical Implications ... 37
5. Conclusion ... 39
References ... 40
5 List of Tables and Figures
Figure 1: Research Model
Table 1: Respondent Characteristics
6 Introduction
“The smartphone revolution is under-hyped, more people have access to phones than access
to running water.” - Marc Andreessen, co-founder of Andreessen Horowitz
Mobile phones have changed rapidly during the last decade, as they have become
more than just a way of verbal communication. Smartphones distinguish themselves from
standard mobile phones by providing continuous internet access and have thereby transformed
into tools that provide virtual environments and digital identities (Gökçearslan, Mumcu, Haşlaman, & Çevik, 2016). By now, over one third of the worldwide population owns a smartphone (Statista, 2016). Smartphones are not only used for social aims (e.g. social media
and WhatsApp) (Pew Research Center, 2015), but also for work-related purposes (e.g.
connectivity to work-related emails and documents) (Lanaj, Johnson, & Barnes, 2014).
Besides the positive facts, like being connected anytime and anywhere, smartphone use also
has it downsides. Previous research on smartphone use has shown that smartphone use
heavily disrupts our work life balance and social activities, it is negatively related to sleep duration, and that overuse can result in smartphone addiction (e.g. Montag, Błaszkiewicz, Sariyska, Lachmann, Andone, Trendafilov, Eibes, & Markowetz, 2015; Lemola,
Perkinson-Gloor, Brand, Dewald-Kaufmann, & Grob, 2015).
For many smartphone users, their phone is often the first thing they look at when waking
up, and the last thing they look at before falling asleep (Lee, Chang, Lin, & Cheng, 2014). This
increase in being online will likely also be visible at work, as boundaries between work and
leisure have become blurry (Kim & Byrne, 2011). However, this possible relationship has not
been given much attention yet. Lim (2002) defined this phenomenon as cyberloafing, a special form of loafing behavior, including any voluntary act of employees’ using their companies’ internet to surf non-job-related websites for personal purposes and to check personal email
7 cost companies a lot of money, it decreases productivity, it can put the organization at risk when
employees for example engage in illegal online activities, and it increases risk of cyber hacking
and distribution of viruses (e.g. Chou, Sinha, & Zhao, 2008; Eastin, Glynn, & Griffiths, 2007;
Panko & Beh, 2002; Zakrzewski, 2016). With so many negative consequences, it is important
to know what can cause cyberloafing. This study will examine whether people who generally
use their smartphones more often after work (for either work-related or social purposes), will
cyberloaf more during work hours the next day.
Next, this study will explore whether people who use their smartphones more often in
the evening, will experience more work-home interference that same evening. Work-home interference can be seen as “a process of negative interaction between work and home domains” (Van Hooff, Geurts, Kompier, & Taris, 2006, p.145). Nowadays people easily take their work
home due to increasing interconnectivity (Derks & Bakker, 2014). As work-home interference
has several negative consequences for individuals like decreases in well-being, reduction of
family life quality, and exhaustion (Kinnunen, & Mauno, 1998), it is important to understand
what causes it.
Previous research on the impact of smartphone use on work-home interference only
focused on work-related smartphone use, which has been mentioned as a limitation of their
research (e.g. Derks & Bakker, 2014). This study will also look at the impact of social
smartphone use, such as personal WhatsApp and social media use, on work-home interference.
As research by Wei and Lo (2006) has shown that smartphone use helps strengthen social
relations, this device might be linked to improvement of well-being and could thus decrease
work-home interference (Park & Lee, 2012). However, social smartphone use does not only
have positive outcomes, as research by Przybylski and Weinstein (2013) found that it can have
negative effects on closeness, connection, and conversation quality, which in turn could
8 Another aim of this study is to examine the possible moderating effect of self-control.
With adequate self-control people are able to inhibit impulses, ignore distracting cognitions and
emotions, adjust behavior with social norms (Lanaj, et al., 2014). Therefore, even when
work-related and social smartphone use is high, people with a high level of self-control will be less
likely to engage in cyberloafing activities, than people with a low level of self-control.
Furthermore, it seems obvious that people who have low levels of self-control will experience
higher levels of work-home interference, as they have less resistance to certain impulses
(Baumeister, Vohs, & Tice, 2007), like regularly checking their phones for work-related
purposes during non-work hours, which in turn could lead to conflicts at home.
Taken together, this research is intended to bring us one step further in finding the causes
for cyberloafing and work-home interference. Although many possible causes for cyberloafing
and work-home interference have been investigated, little attention has been paid to the impact
of work-related vs. social smartphone use after work hours. Since smartphone users are
spending more and more time on their devices (Dunn, 2017), it seems plausible that this is also
prevalent during work hours in the form of cyberloafing and it may lead to conflicts between
work and home domains. These relations will be further investigated by incorporating the
moderating effect of self-control, which could possibly affect those relationships and has not
been given sufficient attention yet.
This paper is structured into five sections. The first section will provide a theoretical
background, including the explanations of the concepts of smartphone use, cyberloafing,
work-home interference, and self-control along with the research hypotheses and the research model.
The second section will explain the research method used to test the proposed hypotheses. In
the third section, results will be discussed. The fourth section will provide a discussion,
including implications, suggestions for future research, and limitations. Lastly, this paper will
9 1. Theoretical Background
1.1 Smartphone use after work hours
A smartphone is a mobile device which almost functions like a pocket PC. It has the functions
to make phone calls, browse the internet, manage the calendar, and to receive and answer emails
anytime, anywhere (Derks, ten Brummelhuis, Zecic, & Bakker, 2014). Pew Research Center
(2016) demonstrated that the world became increasingly interconnected, as the amount of
people in emerging and developing countries who use the internet and own a smartphone has
grown noticeably of the past years. However, a difference with people from advanced countries
remains, as they are still generally more often online and own more high-tech gadgets.
Previous research on smartphone use has shown that smartphone use heavily disrupts our
work life and social activities, and that overuse can result in smartphone addiction (Montag, et
al, 2015). Research by Lemola, et al. (2015) has shown that smartphone use is negatively
related to sleep duration and positively with sleep difficulties, which affects a persons’
wellbeing and leads to depressive symptoms. Furthermore, internet addiction became serious
after launching smartphones. Being able to carry it in the hand and having many available
apps, smartphones are more convenient for the use of internet than PC (Kim, 2013). As
smartphones contribute strongly to internet addiction, this study will investigate whether this
addiction will also be visible during work hours. Current research on smartphone use has
mainly looked at smartphone use in general. Therefore, this research will explore whether
work-related and social smartphone use after work hours will have an effect on cyberloafing
at work the next day. It is possible to restrain employees from using smartphones for
work-related purposes after work hours (by for example setting restrictions), but it is more difficult
to keep social smartphone use of employees after work hours under control. As this
10 this research will investigate whether the results differ when smartphones are used for social
or work-related purposes.
Moreover, smartphone use also leads to blurry boundaries between work and home
domains as this way employees easily take their work home with them (Derks & Bakker,
2014). However, previous research on the impact of smartphone use on work-home
interference focused on work-related smartphone use. According to Derks & Bakker (2014)
staying connected to work via smartphones after work hours leads to being less successful in
initiating activities for recovery in response to high work-home interference. Furthermore,
Sonnentag, and Fritz (2007) showed that people who are able to disconnect from their job
after work hours have higher levels of well-being and satisfaction. Previous research
mentioned that distinguishing between work-related and social smartphone use could be
interesting for further research as they could have different outcomes (e.g. Derks & Bakker,
2014). According to Przybylski & Weinstein (2013) social smartphone use has besides some
positive outcomes also negative outcomes as it can negatively affect closeness, connection
and the conversation quality. Therefore, this study will investigate if work-related and social
smartphone use after work hours actually have different effects on work-home interference
that same evening.
1.2 Cyberloafing
Over the last decades more and more people got internet access and the amount of internet users
increased rapidly (Internet Live Stats, 2016; The World Bank, 2016). Internet became part of
our daily lives, which is also visible at work. Even though, employees have various advantages
of computer-mediated communication, they may also be easier distracted from their work by
the temptation of using internet for personal reasons (Kim & Byrne, 2011). In the literature cyberloafing has been given various names such as ‘cyberslacking’, ‘internet abuse in the
11 workplace’, ‘online loafing’, ‘cyberbludging’, ‘internet deviance’, ‘problematic internet use’, and ‘non-work-related internet activities’ (Kim & Byrne, 2011). However, in this research the term cyberloafing will mainly be used, which Lim (2002) described as special form of loafing behavior, including any voluntary act of employees’ using their companies’ internet to surf non-job-related websites for personal purposes and to check personal email during work time.
Regarding Mills, Hu, Beldona, and Clay (2001), most frequent cyberloafing activities are
gaming and gambling, job hunting, live shows and streaming media, online stock trading,
personal e-mails, perusing pornography and cybersex, recreational surfing, and shopping.
However, during the last decade the use of social media increased enormously, therefore this
can also be seen as a common cyberloafing activity (Pew Research Center, 2015).
Some scholars have argued that personal internet use can have a positive influence on
work productivity as it for example may lead to better task performance (e.g. Coker, 2011).
However, social internet activities during work hours regularly have negative consequences, as
it can cause companies a lot of money, it decreases productivity, it can put the organization at
risk if employees create a harmful environment by engaging in illegal online activities, it
increases risk of cyber hacking, and it can lead to distribution of viruses (e.g. Chou, et al., 2008;
Eastin, et al., 2007; Panko & Beh, 2002; Zakrzewski, 2016). Research by Malachowski (2005)
has shown that employees admit to wasting 2.09 hours per day at work, of which almost half
came from cyberloafing. Griffiths (2003) stated that 59 percent of internet use at work is not
work-related. Therefore, it is in important to understand what causes cyberloafing behavior.
Quite some studies investigated causes for cyberloafing behavior. Research by Eastin,
et al. (2007) has shown that as workplace boredom increases the use of personal internet use at
work also increases. Lim (2002) states that when employees perceive their organization to be
unjust they are more likely to cyberloaf. Moreover, Henle and Blanchard (2008), assert that
12 engage in cyberloafing behavior. They also mentioned that when sanctions to cyberloafing were
uncommon, employees were more likely to cyberloaf in response to the stressors. Kim et al.
(2016) showed with their research that conscientiousness and emotional stability were
negatively related to cyberloafing behavior. Furthermore, a study by Wagner et al. (2012)
showed that cyberloafing can be predicted by a low quality and quantity of sleep the previous
night. Another study has shown that employees who are more habitual users of the internet and
those with a general expectation that internet use will be beneficial to their work, are more
likely to cyberloaf. However, this result was lower for employees who were more committed
to their organization and had more restrictions on the use of work computing systems (Garret
& Danziger, 2008).
A lot of research has been done on the causes of cyberloafing (Kim & Byrne, 2011).
However, the studies related to smartphone use have generalized smartphone use (e.g. Derks &
Bakker, 2014; Derks, Duin, Tims, & Bakker, 2015; Lanaj, et al., 2014). As social smartphone
use has a totally different purpose than work-related smartphone use, this study will distinguish
between these two types of smartphone use. For example, cyberloafing influenced by social
smartphone use could be a reason of fear of missing out (Elhai, Levine, Dvorak, Hall, 2016). Elhai et al. (2016) explains fair of missing out as “a personality construct involving reluctance to miss important information, including social information” (p. 510), which could lead to frequently staying connected to social networks. Moreover, research by Van Deursen, Bolle,
Hegner, and Kommers (2015) revealed that smartphone use for social purposes influences
habitual use, which also could be visible during work hours. On the other hand, (intensive)
work-related smartphone use leads to experiencing more conflicts between work and home
domains, which also leads to more exhaustion. Previous research has shown that people who
are feeling exhausted are more likely to cyberloaf more during work as they see it as an easy
13 study will test the effects of work-related vs. social smartphone use on cyberloafing the next
day at work by testing the following hypotheses:
Hypothesis 1a. Work-related smartphone use after work will be positively related to cyberloafing the next day at work.
Hypothesis 1b. Social smartphone use after work will be positively related to cyberloafing the next day at work.
1.3 Work-home interference
Being able to successfully combine work and private life is a major concern for many
employees. This sometimes creates serious problems or conflicts between work and home
domains, which can also be called work-home interference, work-home conflict, work-family
conflict, or work-life conflict (Boswell & Olson-Buchanan, 2007; Carlson, Kacmar, &
Williams, 2000; Van Hooff, et al., 2006). In this study the term work-home interference will be used, which refers to “a process of negative interaction between work and home domains” (Van Hooff et al., 2006, p.145). This work-home interference can arise in two directions – work
interfering with home (e.g. missing out on family activities due to high amount of time spend
on work activities) and home interfering with work (e.g. not being able to spend enough time
on work activities due to family responsibilities) (Carlson, et al., 2000). As this study focuses
on after-work hours, the work interference with home domains will be addressed.
Several research studies have investigated causes for work-home interference. For
example, research by Nordenmark, Vinberg, and Strandh (2012) found that high levels of job
control and job demands are negatively related to work-life balance. The results of this study
also showed that work-life balance is generally lower for people who are self-employed than
14 vs. employment on work-home interference, and came to this conclusion (e.g. Annink, den
Dulk, Steijn, 2016; Parasuraman, & Simmers, 2001). Furthermore, research by Higgins,
Duxbury, and Johnson (2000) revealed that working part-time was associated with lower
work-home interference, compared to working full-time. This is also proven by Major, Klein, and
Ehrhart (2002), as their research has shown that long work hours are associated with
work-home interference. Moreover, prior research has shown that job stressors, such as work
overload, also have a considerable role in predicting work-home interference (e.g. Ford,
Heinen, & Langkamer 2007; Wallace, 1997).
Recently, the use of smartphones for work in the evening has become a hot topic (Ohly
& Latour (2014). Communication technologies, like smartphones, can make the boundaries
between work and home domains more permeable, as this way people can be connected to their
jobs 24/7 (Derks, et al., 2015). This constantly being connected to work could result into people
carrying around their phones all the time, glancing at them repeatedly, responding to
work-related emails in the evening, which in turn could have a negative impact on their work-life
balance (Middleton & Cukier, 2006). Studies have shown that work-related smartphone use
after work indeed causes work-home interference. However, a recent study by Derks, et al.
(2016) stated that this relationship is not as straightforward as we had previously thought, as
only specific segments of the labor force experienced work-home interference after late night
smartphone use for work-related purposes. This study will investigate this relationship more
thoroughly. Furthermore, this research will distinguish between work-related and social
smartphone use as the effect of social smartphone use on work-home interference has not been
investigated before.
Does having contact with -or social support from- family and friends after work through
social media decrease work-home interference? Research on social support from family and
15 friends can alleviate negative emotional states (Hudson, Elek, & Campbell-Grossman, 2000;
Walen & Lachman, 2000). Leung and Wei (2000) showed that smartphones are used to feel
closer with family and friends and to express care and availability to others. Research by Wei
and Lo (2006) revealed that smartphone use helps strengthen social relations. Smartphones are
often used for maintaining relations, therefore this device might be linked to well-being and
could decrease work-home interference (Park & Lee, 2012). However, according to research
by Kraut, Patterson, Lundmark, Kiesler, Mukopadhyay, and Scherlis (1998) spending more
time online could decline social involvement and increases loneliness, a psychological state
which is associated with social involvement. This in turn could increase work-home
interference. Moreover, research by Przybylski and Weinstein (2013) found that mobile phones
can have negative effects on closeness, connection, and conversation quality. Based on these
contradicting findings, social smartphone use may have benefits for some and disadvantages
for others regarding work-home interference. Therefore, no overall effect of social smartphone
use on work-home interference is expected. This research will test the effects of work-related
vs. social smartphone use after work hours on work-home interference that same evening by
testing the following hypotheses:
Hypothesis 2a: Work-related smartphone use after work will be positively related to work-home interference that same evening.
Hypothesis 2b: Social smartphone use after work will have no effect on work-home interference that same evening.
1.4 Self-control
A factor that could influence the relationship between work/social smartphone use after work
16 capacity for adjusting one’s own actions, particularly to bring them in line with standards, such as values, morals, ideals, and social expectations, and to support the pursuit of long-term goals
(Baumesiter, et al., 2007). With adequate self-control people are able to overcome impulses,
which otherwise could lead to for example procrastination of work (Hagger, Wood, Stiff, &
Chatzisarantis, 2010). Research on self-control says that “acts of self-regulation consume a
resource that is limited, leaving people in a state of ego depletion and making them less able to
exert self-control on a subsequent task” (Job, Dweck, Walton, 2010, p. 1686). Once someone’s
pool of resources is depleted, he/she finds subsequent work activities more demanding and
becomes vulnerable to non-task related distractions and impulses, such as cyberloafing
(Baumeister, et al., 2007).
Previous research has proven that loss of control is central in problematic phone use
(Billieux, Maurage, Lopez-Fernandez, Kuss, & Grifffiths, 2015). Research by Lanaj, et al.
(2014) has shown that smartphone use is associated with depletion. Prasad, Lim, and Chen
(2010) stated that individuals who are unable to show self-control will more likely engage in
counterproductive behaviors, like cyberloafing. Avoiding certain behavior requires
self-control by which someone changes its own behavioral patterns in order to prevent its main
response (Muraven and Baumeister, 2000). An example could be responding to personal
WhatsApp messages instead of performing tasks which need to be completed. The act of not
responding to WhatsApp messages may cost more self-control than responding to WhatsApp
messages. Therefore, from the individuals who regularly spend time on their smartphones in
the evening and also show this behavior in the form of cyberloafing during work hours, the
ones who have high self-control are expected to be better able to restrains themselves than
17 Hypothesis 3a. Self-control will moderate the positive relation between work-related smartphone use after work and cyberloafing the next day at work, such that the
relationship is weaker when self-control is high vs. low.
Hypothesis 3b. Self-control will moderate the positive relation between social smartphone use after work and cyberloafing the next day at work, such that the
relationship is weaker when self-control is high vs. low.
Furthermore, this research will look at the moderating role of self-control on the
relation between smartphone use after work hours and work-home interference. A lot of
research has been done on self-control, but not enough attention has been given on the
moderation of self-control on the relation between late night smartphone use and work-home
interference. It seems obvious that people who lack control or have low levels of
self-control will experience higher levels of work-home interference, as they have less resistance
to certain impulses (Baumeister, et al., 2007), like regularly checking their phones during
non-work hours, which in turn could lead to conflicts at home. Specifically, this study will
examine how self-control moderates the relationship between work-related smartphone use in
non-work hours and work-home interference. The moderating role of social smartphone use
on work-interference will not be tested as no direct relation is expected. Therefore, the
following hypothesis will be investigated:
Hypothesis 3c. Self-control will moderate the positive relation between work-related smartphone use after work and work-home interference that same evening, such that the
18 All in all, this research will focus on the role of work-related vs. social smartphone use
after work hours on cyberloafing at work the next day and work-home interference that same
evening. Besides, the moderating role of self-control on these possible relationships will be
explored. All hypothesized relations are displayed in Figure 1.
19 2. Research Method
2.1 Research Design
For this quantitative research, data is collected via online questionnaires using a diary-study
approach in which respondents filled out a survey twice a day on ten consecutive working
days (Monday till Friday). Participants were asked to fill out the survey as soon as possible
after they received it at the end of the morning and the end of the afternoon. The late morning
surveys were available from 11:00 till 15:00 hours and the afternoon survey from 16:15 until
22:30 hours. After these time zones, surveys were not accessible anymore in order to avoid
invalid responses. The week before the daily surveys, participants were asked to fill out a
one-shot survey which included amongst other questions about demographic characteristics and
self-control. All surveys were available in English and Dutch. The online questionnaires were
administered digitally by four master students and the thesis supervisor.
2.2 Sample and Procedure
The population of this study is people who work at least 32 hours a week in the Netherlands.
Furthermore, they have a typical working day, from approximately nine to five. Respondents
were selected based on non-probability sampling, namely by convenience sampling.
Participants were asked to fill out a survey twice a day during ten consecutive working days,
which was time consuming. Therefore, family members and friends of the students were
approached to participate in this research, as they would likely be committed. Moreover, ten
Bol.com vouchers of 20 euros were raffled under the participants who filled out all the
surveys, as an incentive. The aim was to get at least 60 respondents to ensure the sample was
large enough for analysis. Previous diary studies on smartphone use have shown differences
in response rates, for example research by Lanaj, et al. (2014) showed a response rate of 61%
20 the response rate would be, each of the four students strived for as many participants as
possible, with a minimum of 20 participants each. In total, 84 respondents filled out daily
online questionnaires for ten working days. Characteristics of the respondents can be found in
Table 1. 647 cases were included in the final dataset. The missing cases (193) can be
attributed largely to five respondents who failed to complete the daily questionnaire eight or
more times (38% of total attrition). Missing values were excluded from the analysis.
Table 1: Respondent Characteristics
Characteristics N (%) / M (SD) Gender: Male Female 43 (51.2%) 41 (48.8%) Age (years) 35.06 (SD 12.28)
Work experience (years) 13.24 (SD 12.99)
Working at current employer (years) 7.54 (SD 10.20)
Weekly time spend on work (hours) 41.60 (SD 6.86)
Number of employees responsible for 5.38 (SD 14.17)
2.3 Measures
There were several variables measured in the survey, however only the relevant variables for
this study will be reported. The one-shot survey included questions about demographics, such
as age, tenure (ratio), gender (nominal) and personality aspects such as self-control, dealing
with stress (interval: 5-point Likert-scale). The diary surveys asked participants questions
about smartphone use, cyberloafing behavior, and work-home situations. The questions about
smartphone use were open questions (ratio), which had to be indicated in minutes. For the
questions about cyberloafing and work-home interference a 5-point Likert-scale was used, varying from “not at all – constantly” for cyberloafing and “completely disagree – completely agree” for work-home interference. For the native Dutch participants, the English surveys
21 were translated in to Dutch, to ensure better understanding among the respondents. The thesis
supervisor has checked the questions and translation.
Work-related smartphone use after work
Work-related smartphone use after work was measured by the following question, based on research by Lanaj, et al. (2014): “How many minutes did you use your Smartphone for work
after 9 PM last night?’’.
Social smartphone use after work
Social smartphone use after work was measured by the following question, also based on
research by Lanaj, et al. (2014): “How many minutes did you use your Smartphone for private
purposes after 9 PM last night?”.
Cyberloafing
Cyberloafing was measured based on research by Jia, et al. (2013) by the following question,
“Today, how often did you engage in any of the following [non-work-related Internet activities] for personal reasons? These non-work-related Internet activities included: Check/send e-mail and text messaging, Look up information online such as news, sports scores, or stock reports, Visit social networking sites, Shop online, Download non-work-related information, Play online games.” Jia, et al. (2013) reported a Cronbach’s α of .93.
However, they used the mentioned non-work-related Internet activities as separate items,
whereas in this study they are combined into one item. In this study, cyberloafing was
measured twice a day, once in the late morning (M = 2.14, SD = .93) and once in the late
afternoon (M = 2.32, SD = .87). The two cyberloafing measurements have a correlation of .70,
22 Work-home interference
Work-home interference was measured by the following shortened three item version of
research by Derks, Bakker, Peters, and Van Wingerden (2016): “Yesterday, the demands of
my work interfered with my home and family life”, “Yesterday, the amount of time my job took up made it difficult to fulfill family responsibilities”, and “Yesterday, things I wanted to get done at home did not get done because of the demands my job puts on me”. Cronbach’s α
of the scale was .92 (Derks, Bakker, Peters, & Van Wingerden, 2016). In the current study,
work-home interference was measured in the morning. The three-item scale proved reliable with a Cronbach’s α of .94 (M = 1.91, SD = 0.91).
Self-control
To measure self-control, the following four items based on research by Smit and Barber (2015) were used: “I am good at resisting temptation”, “I have a hard time breaking bad
habits”, “I wish I had more discipline”, “People would say that I have iron self-discipline.” Cronbach’s α of the scale was .75 (Smit and Barber, 2015). In this study, the
scale was acceptable reliable with a Cronbach’s α of .71 (M = 3.14, SD = 0.68). Furthermore, the second and third item on self-control were recoded to let them point in the same direction.
2.4 Analysis
To test the proposed hypotheses, data was collected via a diary study. Participants received a
survey twice a day for ten consecutive working days. Prior to the dairy study, participants
received a one-shot survey which measured personality traits. The data has been analyzed
using IBM SPSS Statistics 23 for Windows. The variable cyberloafing was measured twice a
day (during the late morning and late afternoon). Therefore, there are two variables for
23 morning and T2 refers to the measurement in the late afternoon. Work-related and social smartphone use were centered at respondents’ means of the variable. Centering the items around participants’ means eliminates the effects of between-person confounds, and thereby leads to a better understanding of relations regarding these variables (Lanaj et al., 2014). After
these steps, bivariate correlations were assessed in order to look at the strength of the
relationship between certain variables. Next, multi-level regression analyses were used to
24 3. Results
3.1 Correlations
Before testing the hypotheses, correlations were tested by using the computed means of the
variables shown in Figure 1, including the basic demographic variables age and gender. These
bivariate correlations are displayed in Table 2.
Table 2: Means, standard deviations, and correlations of study variables
1 2 3 4 5 6 7 8
1. Late Night WSU -
2. Late Night SSU -.02 -
3. Cyberloafing T1 .03 .06 - 4. Cyberloafing T2 .06 -0.01 .70** - 5. WHI. .15** .03 -.04 -.05 - 6. Self-control .00 .00 -.02 -.07 -.05 - 7. Age .00 .00 -.28** -.31** .12** -.08* - 8. Gender (F=0, M=1) .00 .00 -.06 -.07 .10* -.02 .17** -
Note: WSU = related Smartphone Use; SSU = Social Smartphone Use; WHI =
Work-home interference.
** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed)
First of all, these results show that late night work-related smartphone use is positively related
to work-home interference (r = .15, p < .001). This means that employees who use their phone
more in the evening for work-related purposes are more likely to experience work-home
interference that same evening. Furthermore, there is a positive relation between work-home
interference and age (r = .12, p < .001), which indicates that older employees are more likely
to experience work-home interference than younger employees. The positive relation between
25 likely to encounter work-home interference than female employees. Independent samples
T-test confirmed that men reported more work-home interference (M = 1.97, SD = 0.63) than
females (M = 1.84, SD = 0.62), t (838) = -3.17, p < .01. Moreover, the correlations indicate
that there is a negative relation between age and cyberloafing measured in the late morning (r
= -.28, p < .001) and late afternoon (r = -.31, p < .001). This means that younger employees
are more likely to engage in cyberloafing than older employees. The negative relation
between control and age (r = -.08, p = .03) shows that younger employees have less
self-control than older employees. Lastly, there is a positive relation between gender and age (r =
.17, p < .001). The distribution between male and female employees was almost equal (49%
female), therefore this result indicates that male in the sample were somewhat older than
females. Independent samples T-test confirmed that men (M = 37.16, SD = 11,88) were
indeed older than women (M = 32.85, SD = 12.54), t (838) = -5.12, p < .001.
3.2 Hypotheses Testing
The first aim was to test whether late night smartphone use was related to cyberloafing the next
day at work. To test hypothesis 1a, a multilevel regression analysis with work-related
smartphone use after work hours as the independent variable and cyberloafing in the late
morning and afternoon as the dependent variables was performed. Results showed that there
was no significant effect of work-related smartphone use after work hours on cyberloafing in
the late morning, B = .001, t (485.21) = 0.76, SE = .002, p = .45 and late afternoon, B = .003, t
(400.69) = 1.47, SE = .002, p = .14. Furthermore, late morning cyberloafing and late afternoon
cyberloafing were summed in order to calculate the total amount of daily cyberloafing. Results
showed that there was also no significant effect of late night work-related smartphone use on
the total amount of daily cyberloafing the next day, B = .004, t (488.86) = 0.94 SE = .004, p =
26 Next, hypothesis 1b was tested in order to see whether there was an effect of social
smartphone use after work hours as the independent variable on cyberloafing in the late morning
and afternoon as the dependent variables. Results showed that there was a marginally significant
effect of social smartphone use after work hours on cyberloafing in the late morning, B = .002,
t (485.18) = 1.66, SE = .001, p = .10 and no significant effect of social smartphone use after
work hours on cyberloafing in late afternoon, B = .000, t (400.69) = 0.304, SE = .001, p = .76.
Furthermore, there was no significant effect of late night social smartphone use on the total
amount of cyberloafing the next day, B = .004, t (488.85) = 1.29 SE = .003, p = .20. This
indicates that when we take a significance of .1, there is an effect of late night social smartphone
on cyberloafing the next morning. However, there is no effect of late night social smartphone
on cyberloafing the next afternoon nor on the total amount of cyberloafing. This means that
hypotheses 1b is partial supported.
The second aim was to test whether late night smartphone use was related to
work-home interference that same evening. To test hypothesis 2a, a multilevel regression analysis
with related smartphone use after work hours as the independent variable and
work-home interference as the dependent variable was performed. Results showed that there was a
significant effect of work-related smartphone use after work hours on work-home interference
that same evening, B = .009, t (491.66) = 4.59, SE = .002, p = .00, which means that
hypothesis 2a is supported. As mentioned in the previous section, the average effect of late
night work-related smartphone use in the regression analysis on work-home interference that
same evening is 1.91 on a 5-point scale. When an employee spends one or more minutes on
his/her smartphone for work-related purposes in the evening, this will lead to an increase in
work-home interference that same evening.
Next, hypothesis 2b was tested in order to see whether there was an effect of social
27 the dependent variable. Results indicated that there was no significant effect of late night
social smartphone use on work-home interference that same evening, B = .001, t (492.01) =
0.86, SE = .001, p = .39. This means that hypotheses 2b is supported.
The last aim was to test the moderating effect of self-control on the relations between
the different variables. This was again done by performing a multilevel regression analysis.
First, hypothesis 3a was tested, which indicates the moderating effect of self-control on the
relation between work-related smartphone use after work hours and cyberloafing the next day.
Results showed that there was no significant moderating effect of self-control on the relation
between late night workrelated smartphone use and cyberloafing in the late morning, B =
-.003, t (8.68) = -.61, SE = .004, p = .56 and late afternoon, B = .000, t (4.97) = -.12, SE = .004,
p = .91. Furthermore, there was also no significant moderating effect of self-control on the
relation between late night work-related smartphone use and the total amount of cyberloafing
the next day, B = .005, t (7.50) = -.59, SE = .008, p = .57. This means that hypothesis 3a is not
supported.
Next, hypothesis 3b was tested, which indicates the moderating effect of self-control
on the relation between social smartphone use after work hours and cyberloafing the next day.
Results showed that there was no significant moderating effect of self-control on the relation
between late night social smartphone use and cyberloafing in the late morning, B = .001, t
(484.23) = .26, SE = .002, p = .79 and late afternoon, B = .003, t (398.23) = 1.24, SE = .002, p
= .18. Moreover, there was also no significant moderating effect of self-control on the relation
between late night social smartphone use and the total amount of cyberloafing the next day, B
= .004, t (5.72) = .87, SE = .004, p = .42. This means that hypothesis 3b is not supported.
Lastly, hypothesis 3c was tested, which indicates the moderating effect of self-control
on the relation between work-related smartphone use after work hours and work-home
28 the relation between late night work-related smartphone use and work-home interference that
same evening, B = .004, t (13.60) = .66, SE = .005, p = .52. This means that hypothesis 3c is
also not supported.
3.4 Additional Analysis
As there is a positive relation between work-home interference and gender (Table 2), which
indicates that male employees are more likely to encounter work-home interference than
female employees, this research investigated whether gender moderates the relation between
smartphone use after work hours and work-home interference that same evening. First, the
moderating effect of gender on the relation between work-related smartphone use in the
evening and work-home interference that same evening was investigated by performing a
multilevel regression analysis. Results showed that there was no significant moderating effect
of gender on the relation between late night work-related smartphone use and work-home
interference that same evening, B = .007 t (8,80) = .92, SE = .007, p = .38. Next, the
moderating effect of gender on the relation between social smartphone use in the evening and
work-home interference that same evening was researched. Again, this was done by
performing a multilevel regression analysis. Results indicated that there was also no
significant moderating effect of gender on the relation between late night social smartphone
use and work-home interference that same evening, B = .003 t (28.23) = .66, SE = .004, p =
.52.
Furthermore, as there is partial support for the relation between late night social
smartphone and cyberloafing the next day, this research also investigated whether the amount
of cyberloafing in the afternoon was influenced by morning cyberloafing when controlling for
late night social smartphone use. Results indicate that social smartphone use after work hours
29 .001, p = .44. There was however an effect of cyberloafing in the late morning on
30 4. Discussion
This research is based on the premise that the world is increasingly interconnected, and that
by now more than one third of the worldwide population owns a smartphone (Statista, 2016).
Current research on smartphone use has mainly looked at smartphone use in general.
However, people have different purposes for smartphones, e.g. social and work-related
smartphone use. It remains largely unknown how these two different types of smartphone use
influence cyberloafing and work-home interference, while these two factors can have various
negative effects on the employer as well as the employee. Addressing this issue, this research
distinguished between the two types of smartphone use, to see whether or not they result in
different outcomes in cyberloafing during work hours and work-home interference that same
evening.
A multilevel regression analysis was performed to answer the research question:
“What is the impact of late night work-related vs. social smartphone use on cyberloafing and work-home interference, and are these effects moderated by self-control?”. Results show that
only late night work-related smartphone use has a positive significant effect on work-home
interference that same evening. No support was found for the relation between late night
work-related smartphone use and cyberloafing the next day. However, there was a marginally
positive significant effect of late night social smartphone use on cyberloafing the next day, but
only in the late morning and not in the afternoon. Contrary to expectations, self-control did
not moderate the relationship of late night social or work-related smartphone use and
cyberloafing or work-home interference. Additional analysis indicated that gender also had no
31
4.1 Theoretical Implications
This research contributed to the existing literature in various ways. The first contribution of
this study is a marginally significant support for the relationship between social smartphone
use in the evening and cyberloafing the next morning. Although employees experience
various advantages of computer-mediated communication, they also seem to be more easily
distracted from their work by the temptation of using Internet for personal reasons (Kim &
Byrne, 2011). This research expected both late night work-related and social smartphone use
to be positively related to cyberloafing during work hours. People who use their phones more
in the evening for work-related purposes could experience more exhaustion during work
hours, which could lead to cyberloafing as an easy way to temporarily take of their mind off
things (Ugrin, et al., 2008). Moreover, people who use their phones often for social purposes
could experience fear of missing out (Elhai et al., 2016) and therefore spend more time on the
web for social purposes during work hours. Result only showed a marginally significant effect
of social smartphone use after work hours on cyberloafing in the late morning the next day,
but no significant effect on cyberloafing in late afternoon and the total amount of cyberloafing
the next day. The marginally significant effect of social smartphone usage after work hours on
cyberloafing in the morning the next day could be due to receiving responses to messages or
emails sent the evening before. Because of these checking habits, employees regularly use
their phones which could lead to subsequent cyberloafing (Oulasvirta, Rattenbury, Ma, &
Raita, 2012). An explanation why there was no significant effect of social smartphone use
after work hours on cyberloafing in the late afternoon the next day, or cyberloafing in general,
could perhaps be found in the fact that employees do experience much stress, which could
result in less time for cyberloafing. The highest level of stress is experienced during the end of
the workday (Dahlgren, Kecklund, & Åkerstedt, 2005), which could therefore result in less
32 effect of work-related smartphone use after work hours on cyberloafing the next day. A
possible reason for this could be that employees do not feel exhausted, which could result in
cyberloafing behavior (Ugrin, et al., 2008). Since there is not a lot of research yet on the direct
relation between smartphone use and cyberloafing, this study provides a solid addition to
existing literature. Most studies investigated other factors that could lead to cyberloafing, like
boredom at work or perceiving the organization as unjust (Eastin, et al., 2007; Lim, 2002).
However, this research showed that social smartphone use in the evening could also cause
cyberloafing the next day (in the morning), which brings us one step further in finding causes
for cyberloafing.
This study also contributes to the literature by investigating the relationship between
social and work-related smartphone use in the evening and work-home interference that same
evening. It was expected that work-related smartphone use after work would be positively
related to work-home interference. As communication technologies, like smartphones, can
make the boundaries between work and home domains more permeable, people can be
connected to their jobs 24/7 (Derks, et al., 2015). Indeed, results indicated that there was a
significant effect of work-related smartphone use on work-home interference. Therefore, this
study adds to the existing literature which stated that communication technologies have a
significant positive impact on work-to-life conflict (e.g. Boswell & Olson-Buchanan, 2007).
Because of contradicting findings, social smartphone use was expected to have no significant
effect on work-home interference. Smartphones are often used for maintaining relations,
therefore social smartphone use might decrease work-home interference (Park & Lee, 2012).
However, spending more time on phones could decrease social involvement and increase
loneliness (Kraut, et al., 1998), thereby diminishing the effect of social smartphone use on
work-home interference. As expected, results indicated that there was no significant effect of
33 relationship between work-related smartphone use and work-home interference, therefore this
study provides a valuable addition to the existing literature as it also incorporated the possible
impact of social smartphone use.
Although self-control plays an important role in counterproductive behaviors, such as
cyberloafing (Prasad et al., 2010), there was no support found within this study for
self-control as a moderator between either late night work-related or social smartphone use and
cyberloafing the next day. It seems possible that other factors may have a stronger moderating
effect on cyberloafing during work hours, such as conscientiousness, emotional stability,
agreeableness or extroversion (Kim et al., 2016; Krishnan, Lim, & Teo, 2010). Furthermore,
self-control also does not moderate the positive relationship between late night work-related
smartphone use and work-home interference. There could be other factors that possibly have a
stronger moderating effect on work-home interference experienced in the evening, such as the
level job demand, the level of job control, or the average working hours (Higgins, 2000;
Nordenmark, et al., 2012). Nevertheless, since self-control does not moderate any of the
possible relationships, this research adds to the existing literature by showing us that
self-control has no moderating effect. Future research could consider other possible moderators.
4.2 Limitations and Future Directions
This study contributes to the understanding of the relationship between social vs. work-related
smartphone and cyberloafing/work-home interference. There are a few limitations. First of all,
a convenience sample was used. Surveys had to be filled out twice a day for ten working days,
which is time consuming for respondents. Therefore, family and friends of the primary
researchers were approached as they would likely be more dedicated to actually fill out the
34 the external validity, this risk was minimized by having four students with different
backgrounds collect the data. Participants varied considerably in age, gender, and work sector.
Another limitation is that late night work-related and social smartphone use were both
measured by only one question. This question was about the amount of time in minutes spent
on using their smartphones for either work-related or social purposes. However, these
concepts were not clearly defined. For instance, someone may see a phone call after work
hours with a colleague which includes both work-related and social purposes as work-related,
whereas others may see it as social, or even make a distinction in minutes spent talking about
work or social issues. Therefore, future research should include more questions on the
purpose of smartphone use to make it more specific, possibly differentiating between different
types of communication. Moreover, this study found a significant positive effect between late
night work-related smartphone use and work-home interference. However, only work-related
activities via smartphones have been researched. It could for example be the case that
someone sees an email coming in via his or her smartphone and uses a laptop to work further
on this. Future research should therefore consider this and not only look at work-related
activities via smartphones, but take all the work-related activities into account.
Another limitation concerns the wording of the questions on work-home interference.
Work-home interference can be experienced in two ways, either work conflicts with
life/family or life/family conflicts with work (Boswell & Olson-Buchanan, 2007). Because
this study focused on smartphone use after work hours there is a focus on conflicts with
life/family after work hours. The level of work-home interference was measured the morning
after the possible experienced work-home interference. However, the questions all referred to
the level of work-home interference experienced the day before, which could be confusing as
this study only focused on work-home interference after work hours. Therefore, future
35 referring to work-home interference in general. This way work conflicts with life/family (e.g.
missing out on family activities due to high amount of time spend on work activities and
life/family conflicts with work (e.g. not being able to spend enough time on work activities
due to family responsibilities) will not be confused.
An additional limitation of this study has to do with late night social smartphone use
and work-home interference. As expected, no significant effect was found. This study
proposed that the positive and negative effects would weigh equal, which leads to social
smartphone use having no direct effect on work-home interference. However, this possible
explanation could not be confirmed by this research. Therefore, future studies could
investigate whether there are effects of social smartphone use are on work-home interference
and which specific factors lead to a possible increase or decrease in work-home interference.
Moreover, future studies could also examine the effects of social smartphone use during work
hours on work-home interference in order to see whether having contact with family/friends
during work hours could decrease experienced work-home interference.
On top of that, this study found a marginally significant effect of late night social
smartphone use on cyberloafing but only in the following late morning. However, it is not
clear why there is only a marginally significant effect in the morning and not in the late
afternoon. This study only measured the amount of cyberloafing in general twice a day,
therefore future research could go more in depth on different types of cyberloafing behavior
and focus more on different times of the workday. This study only looked at the frequency
that people engaged in cyberloafing, but did not consider the amount of time. Therefore,
future research could also consider cyberloafing duration to get a clearer overview of time
spent on cyberloafing.
Research on the effects of smartphone use should continue because these phones play
36 on smartphone use, cyberloafing, work-home interference and self-control can be continued in
different ways. Future studies could for example look at effects of different moderators on the
possible relation between smartphone use and cyberloafing/work-home interference. In this
study there was a focus on the between person differences by looking at the behavior
characteristic self-control. There was no significant moderating effect found, which could
mean that another moderator would have been a better choice. Future research could look at
other behavior characteristics such as conscientiousness, which has already shown its effects
on cyberloafing as well as work-home interference. (Kim et al., 2016; RØVIK, Tyssen, Hem,
Gude, Ekeberg, Moum, & Vaglum, 2007). Therefore, it is a logical step for future research to
investigate if conscientiousness not only shows a direct effect on cyberloafing/work-home
interference, but may also act as a moderator in the relationship between smartphone use and
cyberloafing/work-home interference.
Another option would be to look at possible job characteristics that could moderate the
relationship between smartphone use and cyberloafing/work-home interference. One
possibility would be to look at role conflict. Role conflict can be defined as the extent to
which expectations of a role differ from the actual role (Rizzo, House, & Lirtzman, 1970).
Research has shown that it has an effect on cyberloafing as well as work-home interference
(Greenhaus, Bedeian, & Mossholder, 1987; Henle and Blanchard, 2008). Therefore, it seems
relevant to examine if role conflict, besides having a direct effect on cyberloafing/work-home
interference, also moderates the relationship between smartphone use and
cyberloafing/work-home interference.
Moreover, future studies could explore the relationship between social vs.
work-related smartphone use after work hours and other variables, like depletion. Previous research
has shown that extensive smartphone usage in the evening could lower the quality of sleep
37 work. Thus, it seems interesting to examine if late night smartphone use also has a direct
effect on depletion.
Lastly, future research could make more distinctions between different types of
smartphone use. Besides making a distinction between social and work-related smartphone
use, research could also consider process smartphone use (e.g. Elhai, Levine, Dvorak, & Hall,
2017; Van Deursen, et al., 2015). Where social smartphone use involves personal messaging
and social media, process smartphone involves non-social features engagement, like news
consumption and entertainment such as games (Van Deursen, et al., 2015). Studies have
proven that these different types of smartphone use can result in different outcomes, as for
example Elhaj, et al. (2017) has shown that greater social smartphone use was negatively
related to depression symptoms and process smartphone use was more strongly related to
problematic smartphone usage. Future research could therefore consider process smartphone
use as a separate type of (social) smartphone use to see whether it results in different
outcomes in cyberloafing and work-home interference.
4.3 Practical Implications
This research has two practical implications besides the theoretical implications discussed
before. First of all, this study found that there is a marginally significant positive effect of
social smartphone use in the evening on cyberloafing the next morning. As smartphone use
increases so rapidly and is becoming a habit for many people (Lee, Chang, Lin, & Cheng,
2014) it is important for employers to consider that this behavior could also be visible during
work hours which could have negative consequences. It can for example decrease
productivity of employees and can cost companies a lot of money (e.g. Chou, et al., 2008).
Research has shown that internet filtering and monitoring software for operant conditioning is
38 heavy-handed option could be structuring the work places in a way it enlarges the visibility of
computer-mediated activities. A possible way to do this is by letting computer screens face
the hallways instead of the walls or placing sidewalls so employees cannot see when someone
is approaching their work place (Askew, Buckner, Taing, Ilie, Bauer, & Coovert, 2014).
Future studies could research other possibilities for limiting cyberloafing, as it is probably
easier to limit cyberloafing than social smartphone of employees in their spare time.
Lastly, this research confirmed that there is a significant positive effect of work-related
smartphone use in the evening on work-home interference that same evening. Work-home
interference has various negative consequences for employees as it can decrease well-being, it
can lower the family life quality, and it can lead to exhaustion (Kinnunen & Mauno, 1998).
Therefore, it is important that employers try to balance the experienced work-home
interference of their employees. Employers could for instance set restrictions on log-in times
or the timeframe in which emails can be sent (e.g. only access during 7 in de morning and 8 in
the evening). This way they could make sure their employees mainly engage in work
activities during work hours. Future research could investigate more possibilities for limiting
work-related smartphone use after work hours in order to find new ways to balance
39 5. Conclusion
The world has become increasingly interconnected. More people have access to phones than
access to running water. Constantly being connected via smartphones can have various
consequences on our daily lives. This study explored the effects of late night work-related and
social smartphone use on work-home interference that same evening and cyberloafing the
next day. Furthermore, the moderating effect of self-control on these possible relationships
was investigated. Results showed that work-related smartphone use after work hours had a
positive significant effect on work-home interference that same evening. As work-home
interference can have various negative consequences for employees (e.g. it can lead to
exhaustion or decrease wellbeing) (Kinnunen & Mauno, 1998), it is important to limit
work-related smartphone use after work hours. A possible solution could be setting restrictions on
log-in times. However, future research could investigate other possibilities on how to limit
work-related smartphone use after work hours. Results also indicated that social smartphone
use had a marginally positive significant effect on cyberloafing the next morning.
Cyberloafing has various negative consequences for employers (e.g. it can decrease
productivity of employees and can cause companies a lot of money) (e.g. Chou, et al., 2008)
and should therefore be limited. As it is hard to minimize social smartphone use of employees
in their spare time at night, steps should be taken to limit cyberloafing during work hours.
This could be done by structuring work places in a way that screens are visible. However,
future research could examine other ways to limit cyberloafing. Even though not all expected
relationships were found, the results of the current study extend the already existing literature
40 References
Annink, A., den Dulk, L., & Steijn, B. (2016). Work–family conflict among employees and the self-employed across Europe. Social indicators research, 126(2), 571-593. Askew, K., Buckner, J. E., Taing, M. U., Ilie, A., Bauer, J. A., & Coovert, M. D. (2014).
Explaining cyberloafing: The role of the theory of planned behavior. Computers in
Human Behavior, 36, 510-519.
Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control.
Current directions in psychological science, 16(6), 351-355.
Billieux, J., Maurage, P., Lopez-Fernandez, O., Kuss, D. J., & Griffiths, M. D. (2015). Can disordered mobile phone use be considered a behavioral addiction? An update on current evidence and a comprehensive model for future research. Current Addiction
Reports, 2(2), 156-162.
Boswell, W. R., & Olson-Buchanan, J. B. (2007). The use of communication technologies after hours: The role of work attitudes and work-life conflict. Journal of Management,
33(4), 592-610.
Carlson, D. S., Kacmar, K. M., & Williams, L. J. (2000). Construction and initial validation of a multidimensional measure of work–family conflict. Journal of Vocational
behavior, 56(2), 249-276.
Chou, C. H., Sinha, A. P., & Zhao, H. (2008). A text mining approach to Internet abuse detection. Information Systems and e-Business Management, 6(4), 419-439. Coker, B. L. (2011). Freedom to surf: the positive effects of workplace Internet leisure
browsing. New Technology, Work and Employment, 26(3), 238-247.
Dahlgren, A., Kecklund, G., & Åkerstedt, T. (2005). Different levels of work-related stress and the effects on sleep, fatigue and cortisol. Scandinavian journal of work,
environment & health, 277-285.
Derks, D., & Bakker, A. B. (2014). Smartphone use, work–home interference, and burnout: A diary study on the role of recovery. Applied Psychology, 63(3), 411-440.
Derks, D., Bakker, A. B., Peters, P., & Van Wingerden, P. (2016). Work-related smartphone use, work–family conflict and family role performance: The role of segmentation preference. human relations, 69(5), 1045-1068.
Derks, D., Duin, D., Tims, M., & Bakker, A. B. (2015). Smartphone use and work–home interference: The moderating role of social norms and employee work engagement.
