Boundary role transitions: A day-to-day approach to explain the effects of home-based telework on work-to-home conflict and home-to-work conflict

Abstract

Does working from home on a given day complicate or rather facilitate combining work and home roles that day, why and for whom? To answer these questions, we examined how a teleworking day affects daily work-to-home conflict and daily home-to-work conflict. Based on boundary theory, we expected these relationships to be mediated by daily role transitions and moderated by employees’ preferences to protect their home(/work) domain from work(/home) interruptions. Hypotheses were tested through multilevel moderated mediation modeling using diary data collected during 14 consecutive workdays with 81 employees (N = 678 data points). In line with our expectations, employees were found to make more work-to-home transitions (i.e. interruptions of work activities to deal with home demands during work hours) on teleworking days, which was related to lower work-to-home conflict but higher home-to-work conflict on these days. They also made more home-to-work transitions (i.e. interruptions of home activities to deal with work demands after hours) on teleworking days, which was related to more work-to-home conflict on these days. The latter effect was stronger for employees with a home protection preference. There was no moderating impact of work protection preference. Overall, employees experienced less work-to-home conflict but more home-to-work conflict on teleworking days compared to non-teleworking days.

Keywords

boundary theory daily approach role transitions telework work-home conflict

Introduction

Home-based telework, hereafter referred to as telework, becomes increasingly prevalent in organizations across the world (Allen et al., 2015; Greer and Payne, 2014). This work arrangement allows employees to execute work tasks from home during some portion of the working week using information and communication technologies (ICT) (Bailey and Kurland, 2002). Organizations increasingly invest in telework to address the need of a growing number of employees to combine work with home roles (Greenhaus and Powell, 2003; Kalliath and Brough, 2008). However, to date, it remains unclear whether telework facilitates or rather complicates combining work and home roles (Allen et al., 2015; Butts et al., 2013).

On the one hand, telework gives employees flexibility and autonomy in where and how to work (Kossek et al., 2006). This flexibility may enable employees to better schedule their work hours around their home demands (Golden et al., 2006; Versey, 2015). For instance, on days teleworkers work from home, they can easily interrupt their work activities to do the laundry or to pick up the kids from school on time. Such role transitions on a teleworking day may facilitate fulfilling one’s home roles that day and, accordingly, may lower employees’ work-to-home conflict (Golden et al., 2006), that is, the conflict that occurs when employees’ participation in their work role interferes with their home roles and activities (Greenhaus and Beutell, 1985). In line with this reasoning, several studies have found that telework relates to less work-to-home conflict (Gajendran and Harrison, 2007; Golden et al., 2006).

However, a few other studies have found the reverse effect, that is, that telework is related to more work-to-home conflict (Hammer et al., 2005; Hill et al., 2003; Schieman and Young, 2010). These studies explain this work-to-home conflict enhancing effect by arguing that when employees work from home, their home boundaries are more blurred that day since the home domain is then used for work activities (Duxbury et al., 1992; Schieman and Young, 2010). Because the home boundaries are then more blurred, employees are more inclined to keep thinking of work and keep doing work tasks at home that day, even after hours, which could hinder them to meet their home demands in the evening and, as such, they may experience more rather than less work-to-home conflict that day (Voydanoff, 2005).

Some studies further posit that telework complicates balancing work and home roles because it increases home-to-work conflict (Duxbury et al., 1992; Golden et al., 2006), that is, the conflict that occurs when employees’ home activities and commitments interfere with their work responsibilities (Greenhaus and Beutell, 1985). These studies argue that on days employees work from home, family members more easily interrupt them, even for trivial requests (Allen et al., 2003; Kurland and Bailey, 1999). In addition, the home tasks are more salient while at home, which may trigger employees who are working from home to interrupt their work to address these tasks (Golden et al., 2006). These role transitions may hinder them to fulfill their work role that day, which may induce home-to-work conflict.

Interestingly, all the arguments above refer to daily role transitions—more specifically: role transitions on teleworking days—as an important explanatory mechanism for both the possible conflict-reducing effect and the potential conflict enhancing effects of telework. According to boundary theory (Ashforth et al., 2000), role transitions refer to psychological and/or behavioral switches employees make between the work role and the home role to juggle their work and home responsibilities (Ashforth et al., 2000). On days teleworkers work from home, they are likely to make more work-to-home transitions—that is, interruptions of their work activities to address home demands during work hours. These transitions may facilitate fulfilling one’s home role that day—thereby lowering work-to-home conflict—but could also prevent employees from fulfilling their work role that day—thereby triggering home-to-work conflict. In addition, on days teleworkers work from home, their home boundaries are more blurred and, therefore, teleworkers may make more home-to-work transitions after hours that day—that is, interruptions of home activities to address work demands after hours—which may enhance rather than reduce their work-to-home conflict. Despite the likely importance of role transitions on a teleworking day for understanding the relationship between telework and work-to-home and home-to-work conflict, research on these relationships has not yet included role transitions nor examined these relationships on a daily basis (Bailey and Kurland, 2002; Gajendran and Harrison, 2007). If telework affects employees’ work-home interface via daily role transitions on teleworking days, teleworkers’ and non-teleworkers’ work-to-home and home-to-work conflict may only differ on days teleworkers work from home and not on days they work at the office. Therefore, building on boundary theory, the first aim of this study is to examine whether a teleworking day affects employees’ work-to-home and home-to-work conflict—over and above the impact of being a teleworker—via daily role transitions.

A second aim of this study is to better understand for whom the favorable path (i.e. telework lowering work-to-home conflict) and for whom the unfavorable paths (i.e. telework enhancing work-to-home and home-to-work conflict) are likely to dominate. Research has shown that individual and situational characteristics, such as personality traits (Raghuram and Wiesenfeld, 2004), family demands (Byron, 2005) and job demands (Glavin et al., 2011) moderate the relationship between telework and work-to-home and/or home-to-work conflict. Building on boundary theory, we expect that employees’ segmentation preferences play a crucial moderating role because these preferences affect how stressful role transitions are for employees and, thus, how much conflict these transitions are likely to trigger (Derks et al., 2016; Gadeyne et al., 2018). Segmentation preferences have been shown to be domain specific: employees’ preference to protect their work from home interruptions (i.e. work protection preference) differs from their preference to protect their home from work interruptions (i.e. home protection preference) (Methot and LePine, 2016). These segmentation preferences are considered to be distinct constructs; so, employees may be high (/low) on work and high (/low) on home protection preference at the same time or be high on one preference and low on the other (Methot and LePine, 2016). The stronger employees’ home (/work) protection preference, the more stressful work (/home) interruptions are likely to be (Kreiner, 2006). A recent study for instance showed that boundary violations (i.e. role transitions that violated employees’ work-home boundary more than they preferred) increased employees’ work-to-home and home-to-work conflict (Hunter et al., 2017). We add to this study by arguing that role transitions may not only be stressful but could also be helpful, depending on the type of role transition and on the fit with an employees’ segmentation preference (Ashforth et al., 2000). Accordingly, we expect work and home protection preference to moderate the relations between role transitions and work-home conflict.

The contributions of this study are threefold. First, by examining daily role transitions as mediators in the relationship between telework and work-to-home and home-to-work conflict, our study helps to understand through which mechanisms telework affects the work-home interface. Even more, by including two types of role transitions as a mediator (i.e. work-to-home and home-to-work transitions), we simultaneously account for both favorable and unfavorable effects of telework, which may yield a more nuanced understanding of the impact of telework. Second, by examining the moderating role of segmentation preferences, our study helps to understand for whom the favorable and for whom the unfavorable effects of telework are likely to dominate. Insights into such moderators can help to understand the inconsistent results on the relation between telework and work-home conflict found in research to date (Allen et al., 2015; Butts et al., 2013). In addition, we take into account the directionality of employees’ segmentation preference, that is, whether it concerns employees’ work or their home protection preference (Methot and Lepine, 2016). A few studies examined if segmentation preferences moderate the impact of telework (Hyland et al., 2005; Lapierre et al., 2016) or the impact of role transitions (e.g. Derks et al., 2016; Gadeyne et al., 2018); yet, these studies did not always find effects, perhaps because they ignored the domain-specificity of these preferences. Building on the result of Hunter and colleagues (2017) that daily boundary violations—thus, role transitions from one domain that are not in line with one’s preference to protect that domain—increase daily work-to-home and home-to-work conflict, we test the moderating role of the domain-specific segmentation preference upon the relation between daily role transitions and daily conflict. Thirdly, we look at the relationships between telework and conflict on a daily level since the effects of telework may mainly occur on teleworking days. Even though teleworkers rarely work from home every day, research on telework has rarely applied a daily diary design (for an exception, see the study of Vega and colleagues (2015) who found that teleworkers’ job satisfaction and performance was different on teleworking days than on office days). Using a daily diary design also answers calls to capture role transitions at the day they occur (Smit et al., 2016) and to conceptualize work-home conflict as a daily fluctuating experience (Hunter et al., 2017; Shockley and Allen, 2015). Overall, the results of our study can inform organizations about why and for whom working from home facilitates or rather complicates combining work and home roles.

Theoretical background and hypothesis development

Daily work-to-home and home-to-work conflict and telework

Most studies on work-to-home and home-to-work conflict are cross-sectional studies focusing on employees’ general levels of conflict (Casper et al., 2007). Several scholars have recommended to move away from such a stable-level approach (i.e. using a global judgment of work-home conflict) to a dynamic episodes approach (i.e. using daily judgments) because employees’ conflict experiences are likely to vary daily as a result of daily fluctuating role behaviors (Hunter et al., 2017; Maertz and Boyar, 2011; Shockley and Allen, 2015). Research has indeed found work-to-home and home-to-work conflict to fluctuate on a daily basis (Ilies et al., 2007; Maertz and Boyar, 2011). Surprisingly, research on the relationship between telework and work-to-home and home-to-work conflict to date has exclusively used a stable-level approach. However, teleworkers rarely work from home every day and their role behaviors―and, thus, their conflict―may strongly depend on whether they worked from home that day or not. Moving towards a dynamic episodes approach to understand the impact of telework may therefore improve our understanding of the phenomenology of telework (Maertz and Boyar, 2011; Vega et al., 2015). In line with this rationale, this study examines the relationships between a teleworking day and daily work-to-home and home-to-work conflict. In particular, building on boundary theory, we expect that these relationships are mediated by daily role transitions and moderated by segmentation preferences (see Figure 1). In what follows, we first explain boundary theory and then develop our hypotheses.

Boundary theory and boundary role transitions

Boundary theory posits that individuals create and maintain psychological, physical and/or behavioral boundaries around their different life roles, such as their work and home roles (Ashforth et al., 2000). These boundaries are characterized by a degree of permeability, which refers to the ease with which psychological or behavioral aspects of other life roles can cross these boundaries (Ashforth et al., 2000). The more permeable the boundaries of a certain life role are, the more likely it is that employees make role transitions from that role to their other roles—thus, the more likely employees are to make psychological and/or behavioral switches to their other roles (Ashforth et al., 2000). The permeability of a certain life role—and, thus, the likelihood of role transitions—can vary over time depending on, among others, the place where the role is performed that day (Allen et al., 2015; Voydanoff, 2005). Thus, permeable boundaries imply that employees can be involved in one role (e.g. talking with or thinking about family members, making a dentist appointment) while being located in another role/domain (e.g. at work) (Ashforth et al., 2000). For instance, when employees work from home, their work and home roles are temporarily co-located in the home domain, which may make the boundaries of both roles more permeable that day (Allen et al., 2015). Consequently, role transitions are then more likely to occur (Lapierre and Allen, 2006).

According to boundary theory, role transitions can either reduce or enhance work-home conflict, depending on the type of these transitions. Whereas some role transitions may facilitate combining work and home roles (e.g. interrupting work to pick up the kids), other role transitions (e.g. starting to work again in the evening) may enhance conflict (Ashforth et al., 2000). Boundary theory further suggests that how role transitions relate to work-home conflict is person-specific and may depend on, among others, the extent to which employees prefer their roles to be permeable (Kreiner, 2006; Michel and Clark, 2013). While some individuals prefer impermeable boundaries to keep their different roles as separated as possible (i.e. segmentation preference), others prefer to have more blurred boundaries and to make more transitions between their roles (i.e. integration preference) (Kreiner, 2006). For employees with a segmentation preference, role transitions (i.e. their boundary crossing behavior) are incongruent with their preferred boundary management style (i.e., their boundary crossing preference) and, therefore, these transitions are likely to induce more strain and trigger more conflict (Ashforth et al., 2000; Kreiner, 2006; Michel and Clark, 2013). A few studies have indeed found segmentation preferences to moderate the effect of boundary crossing behaviors (Edwards and Rothbard, 1999; Matthews et al., 2010). Since most individuals do not value each role to the same extent, they may mind interruptions of certain roles more than interruptions of other roles (Maertz and Boyar, 2011). Several researchers have therefore argued (Kossek and Lautsch 2012; Powell and Greenhaus, 2010) and one study has shown (Methot and Lepine, 2016) that employees’ segmentation preference is domain-specific, that is, that employees’ preference to protect the work role from home interruptions (i.e. work protection preference) is different from their preference to protect the home role from work interruptions (i.e. home protection preference). However, to our knowledge, research on role transitions and work-home conflict to date has not yet examined the moderating role of these domain-specific preferences. Knowledge on these moderating effects may explain current observed inconsistent relationships between role transitions and work-home conflict.

Telework and boundary role transitions

Building on boundary theory, we first expect that on days employees are telecommuting, they make more work-to-home transitions than on days they are working onsite. On days employees work from home, they perform their work tasks while being physically located in their home domain (Shumate and Fulk, 2004). Since the home domain is both psychologically and behaviorally associated with employees’ home tasks and responsibilities, the boundaries of the work role are likely to become more permeable while working at home (Voydanoff, 2005). Accordingly, interruptions of the work role to perform some home tasks are more likely (Ashforth et al., 2000). These work-to-home transitions enable employees to address some home responsibilities during work hours, which is precisely one of the main reasons why employees often want to telework (Fonner and Stache, 2012). Research has indeed shown that the co-location of work with home may increase the ability to make—and therefore the occurrence of—work-to-home transitions (Matthews et al., 2010). We therefore expect more work-to-home transitions on teleworking days:

Hypothesis 1: On days teleworkers work from home, they make more daily work-to-home transitions than on days they do not work from home.

Second, we expect that teleworkers experience more home-to-work transitions on teleworking days than on non-teleworking days. On a teleworking day, not only the work role, but also the home role is likely to be more permeable because the physical place that is normally reserved for the home role has now been used to address work role demands during the day. In addition, on days teleworkers work from home, the work day often gets prolonged or extended, for instance because they want to finish up some work tasks or want to compensate for the home tasks done during the work hours earlier that day (e.g. a parent who had stopped working early that day to pick up the children from school may continue working in the evening to catch up) (Greer and Payne, 2014). As such, teleworkers may make more home-to-work transitions after hours on teleworking days than on non-teleworking days:

Hypothesis 2: On days teleworkers work from home, they make more daily home-to-work transitions than on days they do not work from home.

Boundary role transitions and work-home conflict

Next, we expect that work-to-home transitions—that is, interruptions of the work role to address some home tasks during work hours—can reduce work-to-home conflict (Ashforth et al., 2000; Voydanoff, 2005). Making work-to-home transitions allows employees to respond to home demands during work hours, which may facilitate fulfilling one’s home role that day and may therefore decrease work-to-home conflict that day (Voydanoff, 2005). In addition, since work-to-home role transitions can help employees to adjust their work hours to their home tasks, employees are likely to feel more in control of the interactions between their different life domains, which is believed to reduce negative spillover effects from one domain to the other (Versey, 2015). Several studies have found perceived control to be negatively related to work-to-home conflict (Maume and Houston, 2001; Versey, 2015):

Hypothesis 3: Daily work-to-home transitions are negatively related to daily work-to-home conflict.

However, on days when employees make more work-to-home transitions to address their home demands during work hours, they may find it harder to fulfill their work tasks that day and could therefore experience more home-to-work conflict (Golden et al., 2006; Lapierre and Allen, 2006). Work-to-home transitions during work hours may reduce the time that is left to fulfill one’s work demands that day and could mentally distract employees from their work tasks (Lapierre and Allen, 2006). This may result in reduced focus and decreased work task performance that day (Smit et al., 2016). Accordingly, when employees make more work-to-home transitions on a given day, they may feel that their home activities hinder them more to fulfill their work role that day and, thus, they may report more home-to-work conflict:

Hypothesis 4: Daily work-to-home transitions are positively related to daily home-to-work conflict.

In addition, home-to-work transitions—that is, interruptions of the home role to address work tasks outside work hours—are likely to increase work-to-home conflict. The more often employees interrupt what they are doing at home to perform work tasks after hours, the less time and resources they are likely to have available for participating in their home activities, such as care and household tasks or quality time with family members. This may hinder employees in allocating their required amount of time to their home role that day and, therefore, increase work-to-home conflict (Carlson et al., 2015). In addition, home-to-work transitions may hinder employees to mentally detach and disengage from work (Derks et al., 2016; Schieman and Young, 2010), which may also enhance work-to-home conflict (Demerouti et al., 2014; Golden, 2012; Voydanoff, 2005). Two prior cross-sectional studies (Carlson et al., 2015; Matthews et al., 2010) and one daily diary study (Hunter et al., 2017) have indeed found home-to-work transitions to relate positively to work-to-home conflict:

Hypothesis 5: Daily home-to-work transitions are positively related to daily work-to-home conflict.

Mediating mechanisms of boundary role transitions

Our hypothesized pathways between a teleworking day and role transitions (Hypotheses 1–2) and between role transitions and work-to-home and home-to-work conflict (Hypotheses 3–5) imply that we expect the effect of a teleworking day on both work-to-home conflict and home-to-work conflict to be mediated by role transitions:

Hypothesis 6: Daily work-to-home transitions mediate the negative relationship between a teleworking day and daily work-to-home conflict (Hypothesis 6a) and the positive relationship between a teleworking day and daily home-to-work conflict (Hypothesis 6b).

Hypothesis 7: Daily home-to-work transitions mediate the positive relationship between a teleworking day and daily work-to-home conflict.

The moderating role of segmentation preferences

Finally, in line with boundary theory, we expect that the relations between role transitions and work-home conflict depend on employees’ domain-specific segmentation preferences (i.e. their work protection preference and their home protection preference). Employees who have a higher work protection preference (i.e. who want to protect their work role from home interruptions) are likely to experience work-to-home transitions as more stressful because these transitions are incongruent with their work role protection preference (Ashforth et al., 2000) and, therefore, they may experience more conflict as a result of these transitions (Ashforth et al., 2000). As such, the relationship between work-to-home transitions and work-to-home conflict is likely to be less negative and the relationship between work-to-home transitions and home-to-work conflict is likely to be more positive when employees’ work protection preference is higher. Similarly, employees with a higher home protection preference (i.e. who want to protect their home role from work interruptions) may experience home-to-work transitions as more stressful because these transitions are incongruent with their home protection preference (Kreiner, 2006) and, therefore, they may experience more work-to-home conflict as a result of these transitions (Ashforth et al., 2000; Kreiner, 2006):

Hypothesis 8: Work protection preference moderates the relation between work-to-home transitions and work-to-home conflict, such that this relationship is less negative for employees high on work protection preference (Hypothesis 8a), and the relation between work-to-home transitions and home-to-work conflict, such that this relationship is more positive for employees high on work protection preference (Hypothesis 8b).

Hypothesis 9: Home protection preference moderates the relation between home-to-work transitions and work-to-home conflict, such that this relationship is more positive for employees high on home protection preference.

Methodology

Procedure and participants

We collected daily diary data from Flemish employees with parental responsibility, because parents encounter specific work-home challenges (e.g. raising a child) and variance in work-home conflict is high within this group (Madsen, 2003). Employees were recruited through a call for participation sent out by one postgraduate student and four undergraduate students from different regions in Belgium to increase geographical distribution. This resulted in a convenience sample of 81 employees. Convenience samples are considered to be quite suitable for elaborated research designs such as daily diary studies (Demerouti and Rispens, 2014). Fifty-three of our respondents were teleworkers (i.e. employees who telecommute at least one day per week) and 28 were non-teleworkers (i.e. employees who did not telecommute on a regular basis). We included non-teleworkers in our sample to be able to examine whether a teleworking day has an effect over and above the impact of being a teleworker and, if so, whether the former effect is indeed more important than the latter.¹

Data were collected through online questionnaires sent via email. Respondents first filled out a general survey with background information, including demographics and segmentation preferences. Then, respondents received a short daily survey during 14 consecutive workdays. We opted for 14 workdays to ensure that we had multiple teleworking days per teleworker. Respondents were instructed to fill out the survey just before going to bed. Of 93 employees who filled out the first general survey, 81 respondents fully filled out at least one of the daily surveys (response rate = 87%). Daily response rates varied from 30% to 79%, and respondents filled out the daily questionnaire between 1 and 14 times in total (M = 8.81, SD = 3.53), resulting in 678 out of 1134 possible observations (60%).

The majority of the sample was female (65%) and worked full time (76%). Official work hours per week ranged from 18 hours to 40 hours (M = 35.21, SD = 5.16). The sample consisted of 35% professional workers, 21% clerks, 20% middle managers and 25% had another function. Tenure ranged from four months to 24 years (M = 9.03, SD = 5.71). Respondents had one to four children (M = 1.94, SD = 0.78) of which the youngest child was maximum 11 years old (M = 4.56, SD = 3.60). All respondents had a partner.

Of the 53 teleworkers, 29 worked from home on average one day a week, 14 respondents worked from home on average two days a week and 10 respondents worked from home on average more than two days a week. Of the 28 non-teleworkers, 15 said that telework was not available in their organization, ten respondents said that this practice was not available for their function and three respondents indicated it was available but they did not use it. We also ran ANOVA analyses to see whether teleworkers differed from non-teleworkers on any of the study’s variables. Results showed that teleworkers experienced more work-to-home transitions (M = 2.86, SD = 1.43) than non-teleworkers (M = 1.92, SD = 1.21; F(1,79) = 15.68, p < .01) and that teleworkers had more job autonomy (M = 4.71, SD = 0.77) than non-teleworkers (M = 3.59, SD = 1.26; F(1,79) = 25.06, p < .01). We found no other differences between teleworkers and non-teleworkers in our sample.

Trait measures

Work and home protection preferences and several control variables were measured once in our general survey. In line with several other studies (e.g. Kreiner et al., 2009), we consider these preferences as personal characteristics that stay stable over short time spans.

Work protection preference

We measured work protection preference with the four-item scale of Methot and LePine (2016) that measures ‘preference to protect the work domain from home intrusions’. Respondents had to evaluate the items on a seven-point Likert scale (1 = Not applicable at all; 7 = Fully applicable). A sample items is ‘I prefer to keep non-work life at home’. The scale was found to be reliable (Cronbach’s α = 0.84).

Home protection preference

We measured home protection preference with the four-item segmentation preference scale of Kreiner (2006), which was later labeled as ‘preference to protect the home domain from work intrusions’ by Methot and LePine (2016) because of its one-directional nature. Respondents had to evaluate the items on a seven-point Likert scale (1 = Not applicable at all; 7 = Fully applicable). A sample item is ‘I prefer to keep work life at work’. The scale was found to be reliable (Cronbach’s α = 0.86).

Controls

We included gender (0 = woman; 1 = man), age (in years), and number of children as control variables because women, older employees, and employees with more care dependent children have been shown to experience more work-home conflict (Butts et al., 2013; Byron, 2005; Madsen, 2003). In addition, we controlled for job autonomy (measured using the 3-item scale of Hackman and Oldham, 1976; e.g. ‘This job gives me considerable opportunity for independence and freedom in how I do the work’; Cronbach’s α = 0.85), since telework is mostly available in autonomous jobs (Kossek et al., 2006). Job autonomy may affect discretion over role transitions and has been found to be an important predictor of work-home conflict (e.g. Kossek et al., 2006). Finally, we controlled for being a teleworker and for telework frequency. We controlled for being a teleworker (i.e. a dummy variable which is 1 if the employee telecommutes at least once a week and 0 if the employee does not telecommute on a regular basis) to make sure that any effect of ‘a teleworking day’ we find in our analysis does not capture the effect of being a teleworker or not. We also controlled for telework frequency (0 = telecommutes never; 1 = on average one day a week; 2 = on average two days a week; 3 = on average more than two times a week) as this may affect how easily an employee forms scripts that ease telecommuting (Fonner and Stache, 2012) and facilitate making work-home role transitions (Ashforth et al., 2000; Smit et al., 2016).

State measures

Teleworking day, role transitions and work-home conflict were all measured in the daily surveys. We used slightly adapted versions of validated role transition and work-home conflict scales: we rephrased the original items to adapt them to the daily level by adding ‘today’ (for a similar approach: see Ilies et al., 2017) and we replaced ‘family’ by ‘home’ since this allows employees to take into account not only family-related but also other home-related concerns. The latter was done in line with recent calls for broader conceptualizations of the home context in work-home research (e.g. Greenhaus and Kossek, 2014).

Teleworking day

Teleworking day is a dummy variable that is 1 if respondents indicated in the daily survey that they had worked from home that day during the regular work hours, and 0 otherwise. We asked respondents to indicate the situation that was most important for them on that day and specified that working from home meant executing work tasks which are normally done at the work place, at home during regular work hours. Respondents who had not worked that day, for instance owing to illness, were coded as missing.

Daily work-to-home transitions

Daily work-to-home transitions were measured with an adapted version of the work-to-home transition measure of Matthews and colleagues (2010). Respondents were asked to evaluate the following statements on a seven-point Likert scale (1 = Not applicable at all to 7 = Fully applicable): (1) ‘Today, I left during my lunch break to meet home responsibilities’; (2) ‘Today, I interrupted my work to meet a home responsibility (like making a dentist or doctor appointment)’; (3) ‘Today, I answered calls or replied to emails from family members or friends while working’; and (4) ‘Today, I changed the hours I worked to tackle home issues’. We consider the measure to be an index since the four statements may not equally apply every day and, thus, do not necessarily correlate with each other on a daily basis (Bollen and Bauldry, 2011).

Daily home-to-work transitions

To assess daily home-to-work transitions, we adapted four items from Matthews and colleagues’ (2010) home-to-work transition measure. Respondents were asked to assess the following items on a seven-point Likert scale (1 = Not applicable at all to 7 = Fully applicable): (1) ‘Today, I answered to work-related calls or emails outside work hours’; (2) ‘Today, I stopped what I was doing after work hours to call work or to send a work-related mail’; (3) ‘Today, I changed plans at home to meet work-related responsibilities’; and (4) ‘Today, I have gone into work to meet work responsibilities outside work hours’. We consider the daily home-to-work transitions measure to be an index, as the four statements may not equally apply every day.

Daily work-to-home conflict

We measured daily work-to-home conflict using a shortened and slightly adapted version of the Carlson et al. (2000) scale. To decrease respondent fatigue, we used two out of three highest loading items of the strain-based conflict scale and two of the three highest loading items of the time-based conflict scale, resulting in four items. The four items were: (1) ‘Today, I had to miss activities at home due to the amount of time I had spent working’, (2) ‘Today, the time I spend on work responsibilities interfered with my responsibilities at home’, (3) ‘Today, after work, I was so emotionally drained that it prevented me from contributing at home’, and (4) ‘Today, after work, I was too stressed to do the things I enjoy due to all the pressures at work’. Respondents had to indicate to which extent they agreed with the given statements on a scale from 1 (Strongly disagree) to 7 (Strongly agree). The daily Cronbach’s α ranged from 0.76 to 0.92, with an average of 0.85.

Daily home-to-work conflict

We measured daily home-to-work conflict using a shortened and slightly adapted version of the Carlson et al. (2000) scale. We used two out of three highest loading items of the strain-based conflict scale and two of the three highest loading items of the time-based conflict scale, resulting in four items. The four items were: (1) ‘Today, the time spent on home tasks caused me not to spend time in activities at work that could be helpful to my career’; (2) ‘Today, I had to miss work activities due to the amount of time I spent on home tasks’; (3) ‘Today, because I was stressed from responsibilities at home, I had a hard time concentrating on my work’; and (4) ‘Today, tension and anxiety at home weakened my ability to do my job’. Respondents were asked to indicate the extent to which they agreed with the given statements on a scale from 1 (Strongly disagree) to 7 (Strongly agree). The daily Cronbach’s α ranged from 0.70 to 0.97, with an average of 0.84.

Strategy of analysis

We have a two-level model with repeated measurements (daily variables) at the first level (N = 678 occasions) and individuals at the second level (N = 81 respondents). Since we have nested observations (i.e. days nested within employees), we used linear mixed coefficient modeling (MCM) in R. We tested the mediation (i.e. Hypotheses 5–6) using the mediate package in R with quasi-Bayesian Monte Carlo approximation (MacKinnon et al., 2004), which is suitable for parametric multilevel mediation models (Bauer et al., 2006). Effects were computed for each of 10,000 bootstrapped samples, and 95% confidence intervals were computed by determining the effects at the 2.5th and 97.5th percentiles. This method also allows to calculate the percentage of the total effect that is explained by the mediator. To study our cross-level moderation effects (i.e. hypotheses 7–8), we followed the guidelines of Aguinis and colleagues (2013) and grand mean centered level two predictors. We did not person-mean center our level one mediators as they function as both predictor and outcome. Finally, we employed restricted maximum likelihood (REML) estimation, as REML does not expect all fixed effects to be known without errors and maximizes only the portion of the likelihood that does not depend on the fixed effects, which makes this method suitable for complex models including multiple fixed effects (Gilmour et al., 1995), like ours.

Results

Descriptive statistics and multilevel modeling

Table 1 shows the descriptive statistics and correlations of the study’s variables.

Table 1.

Means, standard deviations and correlations among the study’s variables.

Variable	M	SD	1	2	3	4	5	6	7	8	9	10	11	12
1. Gender (1 = female)	0.35	0.48
2. Age	36.53	5.57	.16
3. Number of children	1.94	0.78	.09	.31**
4. Job autonomy	4.84	1.20	.29*	.03	−.17
5. Telework frequency	1.07	1.01	.21	.21	.21	.14
6. Teleworker (1 = yes)	0.65	0.48	.20	.15	.11	.24*	.78**
7. Teleworking day (1 = yes)	0.49	0.35	−.14	.16	.04	−.11	.48**	.31**
8. Work-to-home transitions	2.70	1.04	.11	−.05	.13	.08	.43**	.41**	.12
9. Home-to-work transitions	2.27	1.23	.24**	.24**	.08	.16	.17	.07	.02	.19
10. Work protection preference	3.65	1.35	.01	.05	−.04	−.14	−.02	.08	.04	−.06	.09
11. Home protection preference	4.22	1.44	−.07	−.17	−.13	−.24*	−.28*	−.20	−.15	.01	−.20	.35**
12. Work-to-home conflict	2.18	1.09	−.06	.04	−.05	.00	−.21	−.20	.10	.67*	.40**	.23**	.19
13. Home-to-work conflict	1.79	0.92	.03	−.07	.09	−.09	.03	.06	−.04	.35**	.06	.21	.22*	.48**

p < 0.01, *p < 0.05. N = 81 persons and N = 678 occasions. M = Mean. SD = Standard deviation. Means are on a 1–7 Likert scale, except for age (years), telework and gender (dummies), children (number) and telework frequency (1–3 Likert). Correlations between daily variables are person-mean centered (i.e. based on averaged scores across all measurement occasions per person).

Results of the multilevel moderated mediation models

For both work-to-home conflict and home-to-work conflict, we compared five alternative models: (1) a fixed intercept model with no predictors (i.e. general linear model), (2) a multilevel (i.e. random intercept) model with no predictors (i.e. null model), (3) a multilevel model consisting of only the control variables (i.e. controls only model), (4) a multilevel model consisting of the main effects of both controls and predictors but no interaction effects (i.e. main effects only model) and (5) the full multilevel moderated mediation model that we hypothesized.² For work-to-home conflict, the latter model had a better fit than the four alternative models (see Table 2). However, for home-to-work conflict, the main effects only model was found to have the best fit (see Table 2). Therefore, for hypotheses on home-to-work conflict, we report the findings from the multilevel model without interaction.

Table 2.

Global fit indices and model comparison for the corresponding models using ML estimation.

	d.f.	AIC	BIC	–logLik	Comparison	L ratio
Work-to-home conflict
1. General linear model	2	2420.14	2429.18	1208.07
2. Null model	3	2246.49	2260.04	1120.24	2 vs 1	175.65**
3. Controls only model	9	2254.07	2294.75	1118.04	3 vs 2	4.41
4. Main effects only model	13	2193.73	2252.47	1083.86	4 vs 2	72.76**
5. Full model with interactions	16	2170.16	2242.47	1069.08	5 vs 4	29.56**
Home-to-work conflict
1. General linear model	2	2087.57	2096.61	1041.79
2. Null model	3	1788.10	1801.6	891.05	2 vs 1	301.47**
3. Controls only model	9	1797.53	1838.20	889.77	3 vs 2	2.57
4. Main effects only model	11	1746.20	1795.91	862.10	4 vs 2	57.91**
5. Full model with interaction	13	1747.91	1806.65	860.95	5 vs 4	2.29

p < 0.01, *p < 0.05. N = 81 persons and N = 678 occasions. ML estimation because models with different fixed effects cannot be meaningfully compared using REML estimation (Wood, 2011).

Table 3 shows the results of the multilevel analyses to predict work-to-home transitions (Model 1), home-to-work transitions (Model 2), work-to-home conflict (Model 3) and home-to-work conflict (Model 4). A visual overview of the results is given in Figure 2. As can be seen in Table 2, 69% of the variance in work-to-home transitions, 55% of the variance in home-to-work transitions, 58% of the variance in work-to-home conflict and 48% of the variance in home-to-work conflict is owing to within-person variation. This supports our choice for multilevel analyses. In addition, for teleworkers, 70% of the variance in work-to-home conflict was found to be owing to within-person variation whereas in non-teleworkers this was only 30%. For home-to-work conflict, however, the within-person variance was highly similar for teleworkers (i.e., 46%) and non-teleworkers (i.e. 50%). Telecommuting thus seems to trigger more daily fluctuations in work-to-home conflict than in home-to-work conflict.

Table 3.

Random coefficient modeling results to predict work-to-home transitions (Model 1), home-to-work transitions (Model 2), work-to-home conflict (Model 3) and home-to-work conflict (Model 4).

	Model 1		Model 2		Model 3		Model 4
	Work-to-home transitions		Home-to-work transitions		Work-to-home conflict		Home-to-work conflict
	β	SE	β	SE	β	SE	β	SE
Intercept	2.21**	0.27	2.33**	0.34	2.32	0.30	1.28**	0.25
Gender (0 = male, 1 = female)	0.08	0.24	0.47	0.30	−0.16	0.25	0.09	0.22
Age	−0.03	0.02	0.04	0.03	−0.00	0.02	−0.02	0.02
Children	0.10	0.15	−0.03	0.19	0.04	0.15	0.10	0.13
Job autonomy	0.00	0.10	0.12	0.12	0.10	0.10	−0.04	0.09
Telework frequency	0.06	0.17	0.19	0.22	0.11	0.18	−0.09	0.16
Teleworker	0.32	0.36	−0.48	0.46	−0.38	0.38	0.01	0.33
Teleworking day	1.13**	0.14	0.37**	0.14	−0.58**	0.12	0.13	0.09
Work-to-home transitions (WHT)					−0.09**	0.03	0.15**	0.02
Home-to-work transitions (HWT)					0.24**	0.03
Home protection preference (HPP)					0.08	0.11	0.13	0.07
Work protection preference (WPP)					−0.01	0.10
WHT x WPP					0.02	0.02	NS	NS
HWT x HPP					0.06*	0.02
Variance level 2 (employee)	0.79 (31%)		1.30 (45%)		0.92 (42%)		0.69 (52%)
Variance level 1 (day)	1.81 (69%)		1.60 (55%)		1.28 (58%)		0.62 (48%)

p < 0.01, *p < 0.05. N = 81 persons and N = 678 occasions. NS = model not significant. Age, children, job autonomy, telework frequency, WPP and HPP are centered (grand mean centered). WHT and HWT are not centered (person-mean centered) as they also function as outcomes in the moderated mediation model.

Hypothesis 1 predicted that employees make more work-to-home transitions on a teleworking day than on a non-teleworking day. As can be seen in Table 3 (Model 1), the estimate of a teleworking day on work-to-home transitions (β = 1.13, t = 8.10, p < .01) was significantly positive, supporting Hypothesis 1. Hypothesis 2 predicted that employees make more home-to-work transitions on a teleworking day than on a non-teleworking day, which was also supported by our results (β = 0.37, t = 2.68, p < .01; Table 3, Model 2). Next, we predicted that employees who made more work-to-home transitions on a given day would report less work-to-home conflict (Hypothesis 3) and more home-to-work conflict (Hypothesis 4). In line with these hypotheses, Table 3 shows a negative relationship between work-to-home transitions and work-to-home conflict (β = −0.09, t = −2.74, p < .01; Model 3) and a positive relationship between work-to-home transitions and home-to-work conflict (β = 0.15, t = 6.52, p < .01; Model 4). Hypothesis 5 predicted that more home-to-work transitions on a given day would relate to more work-to-home conflict, which was also confirmed by our results (β = 0.24, t = 7.48, p < .01; Table 3, Model 4).

Hypothesis 6 expected that work-to-home transitions would mediate the relation between a teleworking day and work-to-home conflict (Hypothesis 6a) and home-to-work conflict (Hypothesis 6b). In line with these hypotheses, we found that work-to-home transitions mediated 15% of the total effect of a teleworking day on work-to-home conflict (β = −0.10, CI [–0.18, –0.03], p < .01) and 58% of the total effect of a teleworking day on home-to-work conflict (β = 0.17, CI [0.11, 0.24], p < .01). The quasi-Bayesian Monte Carlo analysis further showed that work-to-home transitions and home-to-work transitions only partially mediated the total effect of a teleworking day on work-to-home conflict as there was still a direct effect on work-to-home conflict (β = −0.58, 95% CI [–0.82, –0.33], p < 0.01; see also Table 3, Model 3: β = −0.58, t = −4.70, p < .01). The relationship between teleworking day and home-to-work conflict was fully mediated by work-to-home transitions as there was no direct effect of a teleworking day on home-to-work conflict (β = 0.13, 95% CI [–0.03, 0.30], p = .12; see also Table 3, Model 4: β = 0.13, t = 1.53, p = .13). The total effect of a teleworking day on work-to-home conflict was found to be negative (β = −0.60, p < .01) and the total effect of a teleworking day on home-to-work conflict was found to be positive (β = 0.31, p < .01).

Hypothesis 7 expected that home-to-work transitions would mediate the relation between a teleworking day and work-to-home conflict. In line with this hypothesis, we found that home-to-work transitions mediated 17% (β = 0.08, CI [0.01, 0.06], p < .05) of the total effect of a teleworking day on work-to-home conflict. The relationship between teleworking day and work-to-home conflict was only partially mediated by work-to-home transitions as there was still a direct effect on work-to-home conflict (β = −0.58, 95% CI [–0.82, –0.33], p < 0.01; see also Table 3, Model 3: β = −0.58, t = −4.70, p < .01). The total effect of a teleworking day on work-to-home conflict was found to be negative (β = −0.60, p < .01).

Hypothesis 8 predicted that for employees who had a high work protection preference, the negative relationship between work-to-home transitions and work-to-home conflict would be weaker (Hypothesis 8a) and the positive relationship between work-to-home transitions and home-to-work conflict would be stronger (Hypothesis 8b). However, we have to reject Hypothesis 8a since the interaction effect in predicting work-to-home conflict was not significant (β = 0.02, t = 0.66, p = .51; Table 3, Model 3), nor was Hypothesis 8b since the model including the interaction effect to predict home-to-work conflict did not show a better fit than the model including only all main effects (χ²(2) = 2.29, p = 0.32; Table 2).

Finally, Hypothesis 9 predicted that for employees with a high home protection preference, the positive relationship between home-to-work transitions and work-to-work conflict would be stronger. Our result show support for Hypothesis 9 since we found a positive interaction effect (β = 0.06, t = 2.55, p = < .05; Table 3, Model 3). As can be seen in the visually plotted effect in Figure 3, the relationship between home-to-work transitions and work-to-home conflict is more positive for employees with a high home protection preference (i.e. one standard deviation above average) than for employees with a low home protection preference (i.e. one standard deviation below average).

Finally, we note that none of the outcome variables was significantly related with whether a person was a teleworker or not―nor with any of the other control variables. This suggests that teleworkers and non-teleworkers make a similar amount of role transitions and experience similar levels of work-home conflict on office days and they only differ on these outcomes on teleworking days. Thus, to understand daily role transitions and daily work-home conflict, a teleworking day seems indeed more important than being a teleworker or not.

Discussion

This study aimed to shed light on how telework affects employees’ work-to-home and home-to-work conflict by using a daily diary approach. Building on boundary theory, we expected that these relationships would be mediated by daily role transitions (Allen et al., 2003; Ashforth et al., 2000) and moderated by employees’ domain-specific segmentation preferences, that is, their home protection and work protection preference (Kreiner, 2006).

Our results supported our hypothesized mediation paths. First, we found that on days teleworkers worked from home, they made more work-to-home transitions, which in turn were related with less work-to-home conflict. This suggests that role transitions can be beneficial for employees’ work-home conflict since they enable employees to respond more easily to their home demands. This result extends earlier studies on role transitions which to date have mainly focused on the costs rather than on the benefits of these transitions (Allen et al., 2003; Ashforth et al., 2000; Hunter et al., 2017). These studies argue that making role transitions has a switching cost, for instance, by depleting self-regulatory resources (Hunter et al., 2017; Smit et al., 2016) or removing resources from one domain to another (Carlson et al. 2015; Matthews et al., 2010), which is expected to increase work-home conflict. Our results show that this is not always the case. More specifically, our results suggest that role transitions may also serve as a resource that facilitates combining multiple roles, since transitioning from the work role to the home role may help to respond to pending home demands (Ashforth et al., 2000; Carlson et al., 2015; Jett and George, 2003).

Second, we found that on teleworking days, employees make more home-to-work transitions after hours, which were related with more work-to-home conflict. Because the home environment becomes the work environment on teleworking days, the boundary between the work and the home domain is likely to get more permeable and more easily crossed (Ashforth et al., 2000; Standen et al., 1999). This may hinder employees to carry out their home role responsibilities (Matthews et al., 2010) and to detach from work in the evening (Olson-Buchanan and Boswell, 2006), increasing daily work-to-home conflict. Yet, our results also showed that the total effect of a teleworking day on work-to-home conflict was negative. So, the work-to-home conflict-reducing effect of work-to-home transitions seems to be stronger than the conflict enhancing effect of home-to-work transitions since overall, employees experience less work-to-home conflict on teleworking days. Yet, since the relationship between a teleworking day and work-to-home conflict was only partially mediated by role transitions, other mediators are important as well. Future research may therefore want to explore other mediators that can account for the conflict-reducing effect.

Third, the increase in daily work-to-home transitions on teleworking days was also related with more home-to-work conflict. This result is in line with earlier findings that have found telework to be positively related with home-to-work conflict (Golden et al., 2006; Lapierre and Allen, 2006). Probably, teleworkers take up more home responsibilities when working from home because the home domain is then more salient (Greenhaus and Powell, 2003), increasing the extent to which home demands interfere with work (Lapierre and Allen, 2006). Since work-to-home transitions fully mediated the relationship between a teleworking day and daily home-to-work conflict, role transitions seem to be a sufficient explanation for the finding in earlier studies that telework is related with increased home-to-work conflict (Golden et al., 2006; Lapierre and Allen, 2006).

We only found support for one of the expected moderation effects, specifically, for the moderating role of home protection preference on the relationship between home-to-work transitions and work-to-home conflict. As hypothesized, home-to-work transitions increased work-to-home conflict for all employees, but this effect was stronger for employees with a higher home protection preference. This result is in line with earlier studies that found employee’s segmentation preference to moderate the effect of segmentation behaviors on work-to-home conflict (Edwards and Rothbard, 1999; Matthews et al., 2010).

We did not find the hypothesized moderating effects of work protection preference on the relation between work-to-home transitions, on the one hand, and work-to-home and home-to-work conflict, on the other hand. This indicates that work-to-home transitions are equally beneficial (i.e. lowering work-to-home conflict) and equally harmful (i.e. increasing home-to-work conflict) for employees with a high work protection preference as for employees with a low work protection preference. Perhaps, we did not find support for the moderating role of work protection preference since most employees in our sample did not score high on this preference (M = 3.65 on a scale from 1 to 7). Another possible explanation could be that work protection preference fluctuates daily, rather than it being a stable characteristic, and that it is this daily preference rather than the stable characteristic that moderates daily effects of role transitions on work-home conflict (van Steenbergen et al., 2018). Future research should further explore this issue.

Theoretical contributions

This study contributes to the literature in several ways. First, we simultaneously accounted for possible positive and potential negative effects of telework on employees’ work-home conflict by including two mediators: work-to-home and home-to-work transitions. In that way, our study extended our knowledge on the mechanisms by which telework affects employees’ daily work-home conflict, which is an important advancement for the literature on telework (Allen et al., 2003, 2015). In addition, by showing that work-to-home transitions are related to lower work-to-home conflict, our study adds to the literature on role transitions. While boundary theory suggests that role transitions can both help and hinder combining work and home roles depending on the type of transitions, research to date has mostly focused on conflict enhancing pathways through boundary blurring. Our results also confirm the earlier suggested importance of distinguishing between work-to-home and home-to-work conflict when studying effects of telework (Golden et al., 2006) and the finding that telework more directly affects work-to-home than home-to-work conflict (Shockley and Allen, 2007).

Second, our findings indicate that the adverse impact of home-to-work transitions on work-to-home conflict depends on employees’ home protection preference. In addition, our finding that home protection preference but not work protection preference had a moderating role supports previous research that stressed the importance of distinguishing between the domain-specificity of employees’ segmentation preference (Methot and Lepine, 2016; Olson-Buchanan and Boswell, 2006; Powell and Greenhaus, 2010).

Third, by studying effects of telework on work-home conflict on a daily basis, we followed up on recommendations from the work-home literature to consider work-home conflict as an episodic event rather than as a general level state (Maertz and Boyar, 2011). Despite evidence that work-home conflict fluctuates substantially from day to day (Hunter et al., 2017; Vega et al., 2015), no research on the relationship between telework and work-home conflict to date had used a daily approach. By zooming in on the daily effects of telework, we helped to clarify how telework affects employees’ daily work-home experiences and shed light on the dynamics of the impact of telework.

Practical contributions

Given the increased prevalence of telework in organizations worldwide (Allen et al., 2015), it is important for managers to understand whether, why and for whom the use of this work arrangement affects employees’ work-to-home and home-to-work conflict—two important outcomes for both employees and organizations because of their known association with life and job dissatisfaction, absenteeism, turnover intentions and lower productivity levels (Allen et al., 2000; Byron, 2005; Gajendran and Harrison, 2007). Our results show that teleworking days enable employees to better fulfill their home roles while working from home. Irrespective of their work protection preference and despite the work-to-home conflict increasing effect of more home-to-work transitions on teleworking days, teleworkers experience on average less work-to-home conflict on days they work from home than on office days. These findings thus support the use of telework as a home-friendly work arrangement. Telework could be made even more home-friendly when the work-to-home conflict enhancing effect of telework via increased home-to-work transitions could be lowered. This could be done, for instance, by a code of conduct about after-hours working, by role modeling of supervisors and by avoiding creating an integration culture, which pushes employees to stay connected to work after hours (Gadeyne et al., 2018). This is at most important as our results show that this work-to-home conflict enhancing effect via home-to-work transitions is worse for employees with a home protection preference and, hence, these employees should be considered in counseling and protected from potential push policies to use telework. Finally, our results show that it is important for organizations to safeguard employees’ performance on teleworking days since employees feel that the work interruptions to address home demands during the day hinder them to perform optimally. Managers may want to avoid this effect, for instance by offering counseling to help employees cope in other ways with pressing home demands while teleworking. Research has shown that employees can benefit from certain psychological techniques to cope with opponent responsibilities from different life roles (Versey, 2015).

Limitations

This research has a number of limitations. First, we measured daily role transitions and daily work-home conflict at the same measurement moment and are thus unable to infer directionality nor causality from these findings. Future research could address this issue, for instance by assessing role transitions in the early evening and the experienced work-home conflict at the end of the day. Second, all measures were self-reported, which may lead to common-method bias (Siemsen et al., 2010). Yet, we person-mean centered the variables as our research questions mainly focused on within-person fluctuations, thereby limiting between-person differences (Ilies et al., 2011) and relatedly, common-method bias problems. Last, we targeted employees with parental responsibilities and, therefore, our findings may not generalize towards employees without children, as research has shown that work-home conflict may be different in nature and, hence, has different antecedents, for these employees.

Suggestions for future research

Researchers have argued to include different dimensions of telework, such as whether the telework is requested by the organization or by the employee (Feldman and Gainey, 1997) or whether the employee has control over when to use telework (Kossek et al., 2006). Since these dimensions affect to what extent work and home interruptions are desired by the employee (Jett and George, 2003), they may influence how telework relates to daily role transitions and/or how role transitions relate to work-home conflict. Relatedly, some research has hinted to look at specific characteristics of role transitions, such as whether the role transitions are self-initiated or other-initiated (Ashforth et al., 2000; Smit et al., 2016). It therefore seems important for future research on this topic to take into account specific dimensions of telework and specific characteristics of role transitions.

Conclusion

This study examined why and for whom working from home affects employees’ work-to-home and home-to-work conflict that day. We revealed work-to-home and home-to work transitions as mediating mechanisms that explain both the positive and the negative impact of a teleworking day on employees’ work-home conflict. Overall, teleworkers experienced less work-to-home conflict and more home-to-work conflict on teleworking days compared to non-teleworking days. We also found support for a moderating role of home protection preference: the more employees preferred to protect their home from work interruptions, the more conflict they experienced from such interruptions. No moderating role was found for work protection preference. Overall, our study supports the use of a daily diary approach to understand the relationship between telework and work-home conflict.

Footnotes

Funding

This research was supported by the Flemish Fund for Scientific Research (FWO) (grant numbers: G063014N and G075419N).

Notes

ORCID iD

Joni Delanoeije

Joni Delanoeije (MSc in Psychology) is a doctoral researcher at the department of Work and Organisation Studies at the KU Leuven, Belgium. Her main research interests include the work-home interface and how new ways of working relate to employee well-being and performance. She is also involved in topics on human-animal interaction, such as including pets and taking into account pet care at the modern (home-)workplace. She has published in Journal of Vocational Behavior and Applied Animal Behaviour Science. [Email: joni.delanoeije@kuleuven.be]

Marijke Verbruggen is Professor at the department of Work and Organisation Studies at the KU Leuven, Belgium. Her main research interests include career transitions, career counseling, flexible work arrangements and work-home balance. She has published in Journal of Organizational Behavior, Journal of Vocational Behavior, European Journal of Work and Organizational Psychology and Human Resource Management Journal, among others. [Email: marijke.verbruggen@kuleuven.be]

Lynn Germeys is a postdoctoral researcher at the department of Work and Organisation Studies at the KU Leuven, Belgium. Her primary research interests center around time-dynamic work-home spillover and crossover mechanisms influencing employees’ and their partner’s well-being and job performance. She has published her research in European Journal of Work and Organizational Psychology, Sleep and Frontiers in Psychology. [Email: lynn.germeys@kuleuven.be]

References

Aguinis

Gottfredson

Culpepper

(2013) Best-practice recommendations for estimating cross-level interaction effects using multilevel modeling. Journal of Management 39(6): 1490–1528.

Allen

Renn

Griffeth

(2003) The impact of telecommuting design on social systems, self-regulation, and role boundaries. Research in Personnel and Human Resources Management 22: 125–164.

Allen

Golden

Shockley

(2015) How effective is telecommuting? Assessing the status of our scientific findings. Psychological Science in the Public Interest 16(2): 40–68.

Allen

Herst

Bruck

Sutton

(2000) Consequences associated with work-to-family conflict: A review and agenda for future research. Journal of Occupational Health Psychology 5(2): 278–308.

Ashforth

Kreiner

Fugate

(2000) All in a day’s work: Boundaries and micro role transitions. Academy of Management Review 25(3): 472–491.

Bailey

Kurland

(2002) A review of telework research: Findings, new directions, and lessons for the study of modern work. Journal of Organizational Behavior 23(4): 383–400.

Bauer

Preacher

Gil

(2006) Conceptualizing and testing random indirect effects and moderated mediation in multilevel models: New procedures and recommendations. Psychological Methods 11(2): 142–163.

Bollen

Bauldry

(2011) Three Cs in measurement models: Causal indicators, composite indicators, and covariates. Psychological Methods 16(3): 265–284.

Butts

Casper

Yang

(2013) How important are work–family support policies? A meta-analytic investigation of their effects on employee outcomes. Journal of Applied Psychology 98(1): 1–25.

10.

Byron

(2005) A meta-analytic review of work-home conflict and its antecedents. Journal of Vocational Behavior 67(2): 169–198.

11.

Carlson

Kacmar

Williams

(2000) Construction and initial validation of a multidimensional measure of work-home conflict. Journal of Vocational Behavior 56(2): 249–276.

12.

Carlson

Kacmar

Zivnuska

Ferguson

(2015) Do the benefits of home-to-work transitions come at too great a cost? Journal of Occupational Health Psychology 20(2): 161–171.

13.

Casper

Eby

Bordeaux

Lockwood

Lambert

(2007) A review of research methods in IO/OB work-family research. Journal of Applied Psychology 92(1): 28–43.

14.

Demerouti

Rispens

(2014) Improving the image of student-recruited samples: A commentary. Journal of Occupational and Organizational Psychology 87(1): 34–41.

15.

Demerouti

Derks

Lieke

Bakker

(2014) New ways of working: Impact on working conditions, work–family balance, and well-being. In: Korunka

Hoonakker

(eds) The Impact of ICT on Quality of Working Life. Dordrecht: Springer, 123–141.

16.

Derks

Bakker

Peters

van Wingerden

(2016) Work-related smartphone use, work-family conflict and family role performance: The role of segmentation preference. Human Relations 69(5): 1045–1068.

17.

Duxbury

Higgins

Mills

(1992) After-hours telecommuting and work-family conflict: A comparative analysis. Information Systems Research 3(2): 173–190.

18.

Edwards

Rothbard

(1999) Work and family stress and well-being: An examination of person-environment fit in the work and family domains. Organizational Behavior and Human Decision Processes 77(2): 85–129.

19.

Feldman

Gainey

(1997) Patterns of telecommuting and their consequences: Framing the research agenda. Human Resource Management Review 7(4): 369–388.

20.

Fonner

Stache

(2012) All in a day’s work, at home: Teleworkers’ management of micro role transitions and the work–home boundary. New Technology, Work and Employment 27(3): 242–257.

21.

Gadeyne

Verbruggen

Delanoeije

De Cooman

(2018) All wired, all tired? Work-related ICT-use outside work hours and work-to-home conflict: The role of integration preference, integration norms and work demands. Journal of Vocational Behavior 107: 86–99.

22.

Gajendran

Harrison

(2007) The good, the bad, and the unknown about telecommuting: A meta-analysis of psychological mediators and individual consequences. Journal of Applied Psychology 92(6): 1524–1541.

23.

Glavin

Schieman

Reid

(2011) Boundary-spanning work demands and their consequences for guilt and psychological distress. Journal of Health and Social Behavior 52(1): 43–57.

24.

Gilmour

Thompson

Cullis

(1995) Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models. Biometrics 51(4): 1440–1450.

25.

Golden

(2012) Altering the effects of work and family conflict on exhaustion: Telework during traditional and nontraditional work hours. Journal of Business and Psychology 27(3): 255–269.

26.

Golden

Veiga

Simsek

(2006) Telecommuting’s differential impact on work-home conflict: Is there no place like home? Journal of Applied Psychology 91(6): 1340–1350.

27.

Greenhaus

Beutell (1985) Sources of conflict between work and family roles. Academy of Management Review 10(1): 76–88.

28.

Greenhaus

Kossek

(2014) The contemporary career: A work–home perspective. Annual Review of Organizational Psychology & Organizational Behavior 1(1): 361–388.

29.

Greenhaus

Powell

(2003) When work and family collide: Deciding between competing role demands. Organizational Behavior and Human Decision Processes 90(2): 291–303.

30.

Greer

Payne

(2014) Overcoming telework challenges: Outcomes of successful telework strategies. The Psychologist-Manager Journal 17(2): 87–111.

31.

Hackman

Oldham

(1976) Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance 16(2): 250–279.

32.

Hammer

Cullen

Neal

Sinclair

Shafiro

(2005) The longitudinal effects of work-family conflict and positive spillover on depressive symptoms among dual-earner couples. Journal of Occupational Health Psychology 10(2): 138–154.

33.

Hill

Ferris

Martinson

(2003) Does it matter where you work? A comparison of how three work venues influence aspects of work and personal/family life. Journal of Vocational Behavior 63(2): 220–241.

34.

Hunter

Clark

Carlson

, et al (2017) Violating work-family boundaries: Reactions to interruptions at work and home. Journal of Management. Epub ahead of print 31 March 2017. DOI: 0149206317702221.

35.

Hyland

Rowsome

(2005)The integrative effects of flexible work arrangements and preferences for segmenting or integrating work and home roles. Journal of Behavioral and Applied Management 6(2): 141–160.

36.

Ilies

Johnson

Judge

Keeney

(2011) A within-individual study of interpersonal conflict as a work stressor: Dispositional and situational moderators. Journal of Organizational Behavior 32(1): 44–64.

37.

Ilies

Liu

X-Y

Liu

Zheng

(2017) Why do employees have better family lives when they are highly engaged at work? Journal of Applied Psychology 102(6): 956–970.

38.

Ilies

Schwind

Wagner

Johnson

DeRue

Ilgen

(2007) When can employees have a family life? The effects of daily workload and affect on work-home conflict and social behaviors at home. Journal of Applied Psychology 92(5): 1368–1379.

39.

Jett

George

(2003) Work interrupted: A closer look at the role of interruptions in organizational life. Academy of Management Review 28(3): 494–507.

40.

Kalliath

Brough

(2008)Work–life balance: A review of the meaning of the balance construct. Journal of Management and Organization 14(3): 323–327.

41.

Kossek

Lautsch

(2012) Work–family boundary management styles in organizations: A cross-level model. Organizational Psychology Review 2(2): 152–171.

42.

Kossek

Lautsch

Eaton

(2006) Telecommuting, control, and boundary management: Correlates of policy use and practice, job control, and work-family effectiveness. Journal of Vocational Behavior 68(2): 347–367.

43.

Kreiner

(2006) Consequences of work-home segmentation or integration: A person-environment fit perspective. Journal of Organizational Behavior 27(4): 485–507.

44.

Kreiner

Hollensbe

Sheep

(2009) Balancing borders and bridges: Negotiating the work-home interface via boundary work tactics. Academy of Management Journal 52(4): 704–730.

45.

Kurland

Bailey

(1999) When workers are here, there, and everywhere: A discussion of the advantages and challenges of telework. Organizational Dynamics 28(2): 53–58.

46.

Lapierre

Allen

(2006) Work-supportive family, family-supportive supervision, use of organizational benefits, and problem-focused coping: Implications for work-family conflict and employee well-being. Journal of Occupational Health Psychology 11(2): 169–181.

47.

Lapierre

van Steenbergen

Peeters

Kluwer

(2016) Juggling work and family responsibilities when involuntarily working more from home: A multi-wave study of financial sales professionals. Journal of Organizational Behavior 37(6): 804–822.

48.

MacKinnon

Lockwood

Williams

(2004) Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate Behavioral Research 39(1): 99–128.

49.

Madsen

(2003) The effects of home-based teleworking on work-family conflict. Human Resource Development Quarterly 14(1): 35–58.

50.

Maertz

Boyar

(2011) Work-family conflict, enrichment, and balance under ‘levels’ and ‘episodes’ approaches. Journal of Management 37(1): 68–98.

51.

Matthews

Barnes-Farrell

Bulger

(2010) Advancing measurement of work and family domain boundary characteristics. Journal of Vocational Behavior 77(3): 447–460.

52.

Maume

Houston

(2001) Job segregation and gender differences in work-family spillover among white-collar workers. Journal of Family and Economic Issues 22(2): 171–189.

53.

Methot

LePine

(2016) Too close for comfort? Investigating the nature and functioning of work and non-work role segmentation preferences. Journal of Business and Psychology 31(1): 103–123.

54.

Michel

Clark

(2013) Investigating the relative importance of individual differences on the work–family interface and the moderating role of boundary preference for segmentation. Stress and Health 29(4): 324–336.

55.

Olson-Buchanan

Boswell

(2006) Blurring boundaries: Correlates of integration and segmentation between work and nonwork. Journal of Vocational Behavior 68(3): 432–445.

56.

Powell

Greenhaus

(2010) Sex, gender, and the work-to-family interface: Exploring negative and positive interdependencies. Academy of Management Journal 53(3): 513–534.

57.

Raghuram

Wiesenfeld

(2004) Work-nonwork conflict and job stress among virtual workers. Human Resource Management 43(2–3): 259–277.

58.

Schieman

Young

(2010) Is there a downside to schedule control for the work-family interface? Journal of Family Issues 31(10): 1391–1314.

59.

Shockley

Allen

(2007) When flexibility helps: Another look at the availability of flexible work arrangements and work–family conflict. Journal of Vocational Behavior 71(3): 479–493.

60.

Shockley

Allen

(2015) Deciding between work and family: An episodic approach. Personnel Psychology 68(2): 283–318.

61.

Shumate

Fulk

(2004) Boundaries and role conflict when work and family are colocated: A communication network and symbolic interaction approach. Human Relations 57(1): 55–74.

62.

Siemsen

Roth

Oliveira

(2010) Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods 13(3): 456–476.

63.

Smit

Maloney

Maertz

Montag-Smit (2016) Out of sight, out of mind? How and when cognitive role transition episodes influence employee performance. Human Relations 69(11): 2141–2168.

64.

Standen

Daniels

Lamond

(1999) The home as a workplace: Work–family interaction and psychological well-being in telework. Journal of Occupational Health Psychology 4(4): 368–381.

65.

van Steenbergen

Ybema

Lapierre

(2018) Boundary management in action: A diary study of students’ school-home conflict. International Journal of Stress Management 25(3): 267–282.

66.

Vega

Anderson

Kaplan (2015) A within-person examination of the effects of telework. Journal of Business and Psychology 30(2): 313–323.

67.

Versey

(2015) Managing work and family: Do control strategies help? Developmental Psychology 51(11): 1672–1681.

68.

Voydanoff

(2005) Consequences of boundary-spanning demands and resources for work-to-family conflict and perceived stress. Journal of Occupational Health Psychology 10(4): 491–503.

69.

Wood

(2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. Journal of the Royal Statistical Society 73(1): 3–36.