The double-edged sword of online access to work tools outside work: The relationship with flexible working, work interrupting nonwork behaviors and job satisfaction

Introduction: Information and communication technologies (ICTs) provide employees with online access to work tools outside work (OAWT), which can be seen as a double-edged sword fostering positive as well as negative aspects of flexibility. In our study, we investigated how OAWT relates to different forms of flexible working, work interrupting nonwork behaviors and job satisfaction.

Method: We used a randomized sample of 758 Austrian employees from a broad range of organizations and tested the hypotheses by means of structural equation modeling.

Results: Our findings revealed that OAWT is associated with available flexibility which relates positively to job satisfaction. However, at the same time, it is associated with required flexibility which relates negatively to job satisfaction and positively to work interrupting nonwork behaviors. OAWT has also been found to strengthen the positive relationship between required temporal flexibility and work interrupting nonwork behaviors, and attenuated the negative relationship between required temporal flexibility and job satisfaction.

Implications: We discuss the practical implications and develop recommendations on how organizations should deal with OAWT.

1. Introduction

With the spread of COVID-19, it was impressively visible how employees' online access to work tools outside work (OAWT; e.g., having access to e-mails or online calendar outside work) helped with the organization of work in numerous industries. Many employees had to work from home, which resulted in odd work hours and access to work outside work hours (1). However, ubiquitous access to work is also accompanied with negative side effects such as mental health deterioration, which was highly likely during the pandemic (2, 3). Since working from home during the pandemic was not an autonomous choice, it is important to investigate the relationship of flexible working and boundary management without the omnipresent fear of COVID-19 (4, 5).

Even prior to the COVID-19 pandemic, information and communication technologies (ICTs) enabled employees to access digitalized information and data storages via mobile devices (6) and work, in particular knowledge work, was carried out anywhere and anytime (7). In general, being able to access work outside work hours grants flexibility, but also increases the likelihood that supervisors ask for availability at odd times (8, 9). Thus, work interrupts nonwork behaviors outside work hours, and flexible working seems to creep into employees' daily practice even without defining it as such. Work interrupting nonwork behaviors, therefore, represent an informal form of flexible working that might not be considered as such (e.g., a call from a colleague is not tracked in the time sheet). Traditionally, flexible working arrangements such as flexitime (i.e., available temporal flexibility) and telecommuting (i.e., available spatial flexibility) are implemented for employees to reconcile paid work and family demands (10, 11). However, being able to work flexibly outside the office also provides ubiquitous access to work even outside work hours. Thus, OAWT might encourage work interrupting nonwork behaviors since employees answer e-mails while commuting or during employees' designated nonwork time (12, 13). Although work interrupting nonwork behaviors have been found to positively relate to work-family conflict (14), there are employees, who favor the integration of both work and nonwork spheres over separating them (15, 16) as it provides flexibility (17) and enables them to individually craft work time and space (18). Nonetheless, integrating work and nonwork bears the risk of impairing workers' wellbeing due to lacking mental detachment from work during private time (13, 19–21).

Individual workers, organizations, and legislation actively need to address this behavioral contemporary phenomenon. Following the requirement that organizations are responsible to protect employees' health and introduce measures ensuring workers' safety and health at work (22), some organization such as Volkswagen have opted for blocking OAWT (23). Up to date, research dealing with this phenomenon mostly investigates the behavioral level. Knowledge about structural prevention measures (e.g., blocking OAWT) are mostly lacking although generally structural prevention measures are considered as more favorable than behavioral prevention measures (24). Therefore, this paper aims to provide an analysis about how OAWT relates to flexible working, work-interrupting nonwork behaviors, as well as job satisfaction. The focus on job satisfaction is mainly because it is widely regarded as the “most focal” employee attitude from the viewpoints of both research and practice and has been found to be strongly related to a wide range of desirable employee outcomes, such as improved job performance, increased life satisfaction and reduced withdrawal behaviors (25) and is associated with mental as well as physical health (26). Therefore, identifying its antecedents is of utmost importance.

To summarize, OAWT can be seen as a double-edged sword providing positive as well as negative affordances and requiring employees to actively manage interruptions from both the work and the nonwork spheres. Since a growing number of jobs and work tasks allow or require flexibility (17, 27), we argue that studies about flexible working need to take into account work interrupting nonwork behaviors. We connect two different research streams: the literature on flexible working and the blurred boundaries between work and nonwork literature. Only recently, due to the COVID-l9 pandemic, both research streams have been started to be researched jointly (5). Both, however, require OAWT and, therefore, we investigate how OAWT shapes flexible working and the relationship with work interrupting nonwork behaviors and job satisfaction (see Figure 1). Going beyond previous research in this field, we explore the particular role of ICT and investigate the access provided by technology, not the usage of technology (28) to derive knowledge about structural parameters relating to different forms of flexibility as well as work interrupting nonwork behaviors and job satisfaction. This enables a more inclusive perspective on contemporary working (29–32), and allows deriving guidelines for organizations and employees on how to deal with ICT-enabled work extension.

FIGURE 1

Figure 1. Overview of research model.

2. Flexible working and job satisfaction

In several organizations, there is a debate about blocking employees' e-mail access outside work (23, 33). Despite its positive effect toward more flexibility (34), endless connectivity to the workplace might impair employees' ability to detach from work while outside of work (35), potentially impairing employee wellbeing and health (5, 13, 20). In this section, we shed light on both positive as well as negative consequences of flexibility and discuss how the bright and the dark sides of flexible working could be related to job satisfaction. Figure 1 summarizes the proposed relationships.

2.1. The bright side of flexible working: Available flexibility

Flexible working encompasses temporal flexibility (36) referring to the variation in the number of hours worked and the timing of work as well as spatial flexibility referring to work outside employers' premises connected to an office infrastructure which often takes place at employees' homes (37). Research about flexible working and job satisfaction rarely considers both temporal and spatial flexibility simultaneously (38). Often, flexible working is in place to reduce negative and stressful spillovers from work to private life (39). This potential for increased reconcilability of work and nonwork is also appreciated by employees and research unequivocally shows that employees' possibility to decide when and where to work positively relates to job satisfaction (40, 41). In fact, regardless of whether the organization formally or informally provides flexible working to its employees, the positive relationship between flexible working and job satisfaction remains (42).

Flexible working enables employees to autonomously decide when and where to work and, thus, grants employees more autonomy and control in their jobs. The job-demands-resources model (43), commonly referred to in work psychology, considers autonomy as a job resource that triggers motivation and satisfaction. In line with that, a lot of studies show the positive relationship between autonomy and psychological wellbeing (44–48). Thus, we hypothesize that the availability of flexible working positively relates to job satisfaction. Following Allen and colleagues (38), we distinguish between temporal as well as spatial flexibility.

Hypothesis 1. Available (a) temporal and (b) spatial flexibility relate positively to employees' job satisfaction.

2.2. The dark side of flexible working: Required temporal flexibility

In addition to employees' flexibility via ICTs, mobile technology users also perceive pressure to stay available and connected to work (5). This paradoxical situation was coined as the “empowerment-enslavement paradox” (49). Similarly, the “autonomy paradox” (50) describes once employees are able to respond to emails outside the office, their usage of mobile e-mail devices leads to an escalation of work-related e-mail behavior over time and they deliberately increase work beyond work hours due to the felt obligation to be available. Despite employees' autonomy to decide on the time to use their email devices for professional communication, they end up using it all the time, which actually diminishes their autonomy in practice. Thus, the autonomy paradox reflects the presence of tension between employees' personal autonomy and felt obligation toward the job and colleagues. Furthermore, when e-mails are regularly answered outside work hours, expectations of fast responses emerge leading to lower autonomy on the long run.

The critical perspective of workplace flexibility is also rooted in sociological work (51) addressing potential risks of the transformation toward increased workplace flexibility. We address the risk of felt obligations to be available via ICTs outside work hours (8) and investigate required temporal flexibility, which refers to employees individually perceived expectations concerning their flexibility in time (52–54). Consequences of workplace flexibility should vary depending on whether the control lies within the employer or the employee (55). Thus, we investigate the relationship between employees' required temporal flexibility and their job satisfaction. Following the job-demands-resources model (43), we consider the requirement as job demand and propose that job satisfaction is impaired when employees are required to be flexible in work time since employees' autonomy is limited.

Hypothesis 2. Required temporal flexibility relates negatively to employees' job satisfaction.

2.3. OAWT relates to the bright and dark side of flexible working

Empirical studies dealing with professional ICT usage either draw on qualitative research methods (34, 49, 50) or investigate the behavioral aspects of ICT-enabled work extension (1, 4, 13, 21). But showing this behavior and responding to professional e-mails outside work, first, requires the structural condition to being able to access work. This possibility was provided ad-hoc for many employees during the COVID-19 pandemic for working from home. However, which side effects accompany the technological possibility to access work from anywhere? In contrast to previous studies, we focus on the access provided by ICT and not on the actual usage of ICTs and investigate how different forms of flexibility relate to employees' online access to work tools outside work (i.e., OAWT).

OAWT has been controversially discussed (13, 52, 56): On the one hand, OAWT provides autonomy and flexibility for employees as it enables them to access work independently from their work hours and working place (7, 28). But on the other hand, following the autonomy paradox (50), employees' professional use of ICT devices outside work fosters a perceived obligation of availability over time. This encourages employees to be available for work purposes also outside work hours leading to the interruption of nonwork behaviors and recovery from work. Thus, we consider OAWT as double-edged sword fostering positive as well as negative aspects of flexibility.

Hypothesis 3. OAWT relates positively to (a) available spatial flexibility, (b) available temporal flexibility, (c) required temporal flexibility, and (d) work-nonwork interruptions.

2.4. OAWT moderates the dark side of flexibility

OAWT enables employees to deliberately decide when to answer e-mail requests. Therefore, we argue that the high convergence of work time and workplace due to OAWT also provides some form of boundary control. The boundary between work and nonwork is described in boundary theory (57). This boundary must be crossed each workday and can be marked by physical (i.e., specific settings), temporal (i.e., working times) as well as psychological aspects (i.e., role identities). Boundary control is defined as “psychological interpretations of perceived control over one's boundary environment” (14) and was found to mediate the negative relationship between ICT demands and rumination (58).

The job-demands-resources model (43) explains how job resources such as autonomy buffer the detrimental effects of high job demands (59, 60). Thus, we assume that the negative relationship between required temporal flexibility and job satisfaction might be attenuated as OAWT provides more autonomy for employees about their boundary between work and nonwork.

Hypothesis 4. OAWT attenuates the negative relationship between required temporal flexibility and job satisfaction.

Over time, perceived obligations of availability outside work emerge merely due to the possibility to access it also outside the office (50). Thus, OAWT results in behavioral consequences such as work interrupting nonwork behaviors. Work-nonwork boundary management strategies (14) build on Greenhaus and Beutell's (61) behavior-based “work-family conflict” model, but do not contain an evaluation of whether the interruptions cause any strain or not (14) as some employees prefer the integration of work and nonwork over separating both spheres (15, 16). Thus, work interrupting nonwork behaviors can be seen as a behavioral consequence of the possibility and/or requirement of flexible working.

Since we are particularly interested in how OAWT moderates the dark side of flexibility, we aim to shed further light on how OAWT shapes the direct relationships between required temporal flexibility and work interrupting nonwork behaviors as well as job satisfaction. Generally, requirements from work might interrupt nonwork activities due to the need for temporal flexibility even when employees do not have OAWT. For example, employees might be asked to work longer hours or change their work hours on short notice. However, due to the affordances of ICT to easily engage in ad hoc boundary crossings between work and nonwork, temporal requirements might have become independent from the workplace and thus, the restraining factor of geographical distance in order to accomplish the perceived obligation vanishes. Therefore, the physical boundary from nonwork to work does not need to be crossed in order to behaviorally act in the work role (62) and employees' additional effort for work interrupting nonwork behaviors is lower when OAWT is available. Also drawing on the autonomy paradox (50), we expect that OAWT strengthens the assumed positive relationship between required temporal flexibility and work interrupting nonwork behaviors.

Hypothesis 5. OAWT strengthens the relationship between required temporal flexibility and work interrupting nonwork behaviors.

3. Method

3.1. Sample

In Austria, each employee is an obligatory member of the Chamber of Labor. We used the Chamber's member list from selected industries in Lower Austria (i.e., wholesale, information and communication, financial and insurance activities, management consultancy, architectural and engineering activities, scientific research and development, and advertising and market research, education). The Chamber of Labour of Lower Austria randomly selected 10,106 employees and sent out paper-pencil questionnaires including return envelopes (postage paid upon reception). We opted for a paper-pencil study to avoid selection bias based on e-mail access. In total, 757 questionnaires were returned. This represents a response rate of 7.5 percent. Although this response rate seems to be very low at first sight, one must consider that many employees were invited, who have no affinity to research and maybe some of them even do not understand German. Our study, unlike most other studies in this area, addressed a wide spectrum of employees targeting the general workforce. Since the response rate is like other studies (63), we consider it as adequate.

Overall, the number of participants was balanced regarding gender (48% male). Nearly half of the participants (42%) had at least one child living in their households. Mean age was 39.97 years (SD = 10.84) and actual work hours (including overtime) amounted to 39.01 hours per week (SD = 10.04) on average. Regarding participants' education level, a third (34%) indicated to have a university or college degree and a further third (33%) a high school diploma as highest completed degree. More than a quarter of the participants (28%) were in a leadership position. On average, employees had been employed for 9.80 years (SD = 9.31) at their organization.

3.2. Non-response bias

Due to anonymous participation in our survey, we cannot compare our sample with non-respondents (i.e., members who were invited to participate, but did not respond). However, we can compare (very limited) basic socio-demographic values of our sample with the members who were invited. In line with existing research (64), the likelihood to participate in the survey was higher among women than among men. Although the percentage of men is higher in the working population (58%), as noted before, it is rather balanced in our sample (48%). Furthermore, the number of white-collar workers is considerably higher in our sample (90%) than in the group of employees invited (68%). Regarding age, our sample reflects the group of employees invited (M = 39). Overall, due to the randomized selection, this study also includes lower skilled employees, who potentially do not have OAWT and are generally under-researched in this area of research (65). We consider the randomized invitations as a strength because it enables us to attract employees who are underrepresented in research and in contrast to most of the studies in this area that rely on nonrandom samples, our results could be generalized.

3.3. Measures

Unless otherwise stated, all items were presented on 7-point Likert scales ranging from strongly disagree (1) to strongly agree (7). Table 1 provides all the items and their factor loadings.

TABLE 1

Table 1. Factor loadings of each item.

3.3.1. Available temporal and spatial flexibilty

We used the German flexible working scale from Gerdenitsch et al. (66). Available temporal flexibility was assessed with three items (Cronbach's α = 0.90) focusing on employees' opportunity to autonomously decide when to work. Available spatial flexibility was also measured with three items (Cronbach's α = 0.81) asking about perceived autonomy to decide where to work and whether online participation in meetings is also allowed.

3.3.2. Required temporal flexibility

We measured required temporal flexibility using three items from the German flexible working scale developed by Gerdenitsch et al. (66) (Cronbach's α = 0.81). The items describe employees' response to perceived organizational requirements regarding working time schedules (53).

3.3.3. Frequency of OAWT

In order to capture OAWT, we developed a three-item scale (Cronbach's α = 0.81) assessing the frequency with which employees have access to their e-mails, work calendar, and work-related software/programs outside their work. Responses to the items ranged from never (1) to always (7). Although it is likely that most employees have OAWT either “never” or “always,” we opted for this answering format to have more information and capture varying access due to situations such as blocked e-mail access during specific times, lacking internet coverage at certain places, using a professional mobile internet device which is not used for private purposes, settings on the mobile phone, etc. As expected, the responses showed a bimodal structure toward the ends of the scale (Md = 5). To avoid information loss, we entered the 7-point values in the analysis but opted for an MLR (maximum likelihood estimation robust to non-normality) estimation, which, as the name suggests, is robust to non-normality.

3.3.4. Work interrupting nonwork behaviors

The three items used are translated from the boundary management scale (14) to investigate work interrupting nonworking behaviors (Cronbach's α = 0.74). It captures employees' approach to demarcate their boundary between work and nonwork by assessing how often employees interrupt their nonwork behaviors due to work issues. Response options ranged from never (1) to always (7). For translating the items, we used Brislin's (67) back-translation method.

3.3.5. Job satisfaction

We used a single item to capture a global rating of job satisfaction (“In general, how satisfied are you with your work?”). Using a single item for job satisfaction is considered a common and adequate procedure in work and organization research (47, 68–71) and has shown acceptable reliability and face validity (72, 73).

3.3.6. Control variables

In organizations, flexible working is often provided to facilitate reconciability of work and family demands, in particular when children have to be cared for, which might be influenced by stereotypical gender roles. Thus, and in line with prior studies in this context (74, 75), we included the number of children in the household, gender (0 = male), age, actual weekly work hours, tenure, and leadership position as control variables in our analysis. Furthermore, the degree of flexible working is contingent on the type of job. To control for the influence of the nature of the job, we asked participants to indicate their job title. We then coded participants' job titles into seven categories based on the International Standard Classification of Occupations (ISCO) classification structure: Managers (n = 67), professionals (n = 67), technicians and associate professionals (n = 78), clerical support workers (n = 187), service and sales workers (n = 225), craft and related trade workers (n = 26), elementary occupations (n = 31), and plant and machine operators (n = 7). Finally, we also included life satisfaction due to its strong relationship with job satisfaction (76). In line with literature (77), life satisfaction was measured with one item on a 7-point scale asking “In general, how satisfied are you with your life?”.

3.4. Analysis

Following Anderson and Gerbing's (78) recommendations, we first assessed the validity and reliability of our measures using confirmatory factor analysis (CFA), and then tested the hypothesized relationships with structural equation modeling (SEM) using Mplus (Version 8) (79). We assessed model fit using the indices recommended by Williams, Vandenberg and Edwards (80). In particular, we used the comparative fit index (CFI), the root mean square error of approximation (RMSEA), and the standardized root mean square residual (SRMR). Good fit is attained when the CFI is above 0.95, the RMSEA is < 0.08, and the SRMR is < 0.10 (80).

One of the key limitations of traditional techniques of analysis of interaction effects, such as moderated regression with observed variables, is that these techniques suffer from low power because they do not control for explanatory variables measurement errors. Therefore, latent interaction modeling with SEM has been proposed by researchers as a better alternative (81, 82). A key advantage of using latent variables and SEM is the ability to control for different types of random and non-random measurement errors. This should in turn result in more accurate parameter estimates (81, 82).

3.4.1. Testing validity and reliability of scales

The factor structures of all scales were tested using CFA. A good model fit was achieved [χ² [131] = 386.36, p < 0.001, RMSEA = 0.06, SRMR = 0.05, CFI = 0.96]. The error terms for access to e-mails and calendar were correlated. It is reasonable that when having access to e-mails, the access to the work calendar is highly connected (as often the same mail programs are used for that). To check whether our proposed factor structure is the most likely one, we compared it with competing models such as two, three, four or five-factor models. Table 2 shows that our proposed six-factor model fits the data best.

TABLE 2

Table 2. Measurement models comparison.

The composite reliability scores for all constructs were above 0.75 and the average variance extracted was above 0.50 as can be seen in Table 3. Composite reliability is the extent to which a set of indicators share in their measurement of a construct (83). It is a measure of the homogeneity and internal consistency of the items that form a scale. Constructs that are highly reliable are the ones in which indicators are highly intercorrelated because this suggests that they are all measuring the same latent construct. Composite reliability values of 0.6 or more are generally regarded as acceptable (84). Therefore, all constructs in our study had high internal consistency (85). The square root of the average variance extracted of each construct exceeded the corresponding inter-construct correlations (85). Therefore, discriminant validity was also achieved.

TABLE 3

Table 3. Means, correlations and reliability.

Since the measures for all variables were rated at the same time and by the same respondents, the likelihood of common method bias affecting the results is high. Therefore, we tested for common method bias using the latent method factor approach following the procedure described by Liang et al. (86). All items were allowed to load on their theoretical constructs and a latent method factor. Average variance extracted by the common method factor was 0.23, which is lower than the 0.50 threshold suggestive of a substantive construct (85). Accordingly, we conclude that common method bias was unlikely to be problematic.

3.4.2. Hypotheses testing

We tested our hypotheses using MLR. The first model tested all associations indicated in Figure 1 without interactions (H1 to H4). The proposed structural model fit was good [χ² [138] = 476.30, p < 0.001, RMSEA = 0.06, SRMR = 0.07, CFI = 0.94]. To keep the model as parsimonious as possible we excluded the non-significant control variables and retained life satisfaction (β = 0.39, p < 0.001), age (β = 0.10, p < 0.01), and leadership position (β = 0.08, p < 0.05) in the model. The more the respondents were satisfied with their life in general, the more they were satisfied with their jobs. The older the respondents were, the slightly higher they indicated to be satisfied with their jobs. Employees holding a leadership position were more satisfied with their job than employees who did not hold leadership positions. Overall, in this model, the predictor variables explained 26% of the variance in job satisfaction (R² = 0.258).

The results about available flexibility were in line with H1a and H1b: Available temporal flexibility (β = 0.15, p < 0.01) and available spatial flexibility (β = 0.19, p < 0.01) both related positively with job satisfaction. In line with H2, required temporal flexibility (β = −0.20, p < 0.001) related negatively with job satisfaction. Required temporal flexibility also related to work interrupting nonwork behaviors (β = 0.37, p < 0.001). OAWT was associated with available temporal flexibility (H3a: β = 0.43, p < 0.001), available spatial flexibility (H3b: β = 0.74, p < 0.001), required temporal flexibility (H3c: β = 0.49, p < 0.001) and work interrupting nonwork behaviors (H3d: β = 0.51, p < 0.001), but not with job satisfaction (β = −0.13, p = 0.10). A positive indirect relationship of OAWT on job satisfaction was found via available temporal flexibility (β = 0.06, p < 0.01) and available spatial flexibility (β = 0.14, p = 0.01) and a negative indirect relationship of OAWT on job satisfaction was found via required temporal flexibility (β = −0.10, p < 0.001). Table 4 provides an overview of the results.

TABLE 4

Table 4. Model testing results.

In the second model, we tested whether OAWT moderates the relationship between required temporal flexibility and job satisfaction controlling for all other relationships indicated in the first model. The interaction effect testing H4 was significant (β = 0.09, p < 0.05). Once OAWT is taken into account, the negative relationship of required temporal flexibility on job satisfaction is attenuated (see Figure 2). The simple slope analysis shows that when OAWT is high (1 SD above the mean), the negative relationship of required temporal flexibility with job satisfaction becomes statistically insignificant (b = −0.015, p = 0.865) whereas when OAWT is 1 SD below the mean (b = −0.320, p < 0.001) and at the mean (b = −0.167, p < 0.001) the relationship between required temporal flexibility and job satisfaction was found to be negative. The predictor variables explain 26% of the variance of job satisfaction (R² = 0.260) in this model.

FIGURE 2

Figure 2. Latent interaction of required temporal flexibility and OAWT on job satisfaction.

Finally, in the third model, we tested whether OAWT moderates the relationships between required temporal flexibility (H5) and work interrupting nonwork behaviors controlling for all other relationships indicated in the first model. Our results (see Figure 3) revealed an interaction effect (β = 0.31, p < 0.001). The simple slope analysis shows that when OAWT is low (1 SD below mean), the relationship between required temporal flexibility and work interrupting nonwork behaviors is negative (b = −0.109, p < 0.05) whereas when OAWT is high (1 SD above mean) the relationship between required temporal flexibility and work interrupting nonwork behaviors is positive (b = 0.567, p < 0.001). The predictor variables explain 74% of the variance of work interrupting nonwork behaviors (R² = 0.736) in this model.

FIGURE 3

Figure 3. Latent interaction of required temporal flexibility and OAWT on work interrupting nonwork behaviors.

4. Discussion

In our study we reflect on the debate about blocking employees' e-mail access outside work (87) and investigated how OAWT relates to available and required flexibility, work interrupting nonwork behaviors and job satisfaction. By considering the role of OAWT and assuming positive as well as negative affordances, we connect the literature of flexible working with ICT-enabled blurred boundaries and contribute in several ways.

First, we bring together two streams of research which are traditionally studied separately. Due to the affordances given by contemporary ICTs, flexible working does not necessarily require formal arrangements (42) as employees can access their work also outside work. Traditionally, studies on flexible working refer to a concept (either researching schedule control or telecommuting), which does not incorporate modern ICTs and the changed nature of flexible working (7). Also, during the COVID-19 pandemic flexible working was not always an autonomous choice since remote work was obligatory for many workers in order to prevent contagion with the virus (5). Since the pandemic, the (non-) academic discourse has emphasized that flexible working increases the likelihood of blurred boundaries and work intensification (88). Thus, despite OAWT's advantages for work enabling employees for more flexibility, a threat of constant availability is likely. When employees respond to a quick professional question via e-mail outside their work hours and outside their office, work interrupts nonwork behaviors independent from the location. However, this does not have to be considered as negative in any case (89). With blurred boundaries between work and nonwork, the conceptualization of work itself undergoes a fundamental shift toward more flexibility (34) and the traditional conceptualization of flexible working needs to be re-thought. Future studies need to further investigate how much flexibility is positive and when does too much flexibility turn into detrimental (health) effects?

Second, studies about professional ICT-usage after work hours (35, 90, 91) emphasize the potential detrimental effect on wellbeing due to the lack of mental detachment from work. We extend the behavioral perspective of extent of usage and investigate the structural dimension to derive conclusions about how organizations should deal with OAWT. Our results revealed that OAWT can indeed be considered as a double-edged sword. OAWT is positively associated with the beneficial aspects of flexibility such as available flexibility as well as with the detrimental aspects of flexibility such as required flexibility. Thus, when implementing flexible working, it is crucial to consider potential negative aspects as well as the positive ones.

Third, we show how OAWT plays a role for the relationships on the “dark side of flexibility.” Only when OAWT is high, the relationship between required temporal flexibility and work interrupting nonwork behaviors is positive. When OAWT is low, the relationship is negative indicating that some perceived obligation might not be fulfilled on the behavioral level. This might suggest that blocking OAWT is a suitable measure for organizations to prevent employees from exhaustion. However, blocking OAWT would also limit employees' boundary control since our results also revealed that OAWT helps to cope with required temporal flexibility attenuating the negative relationship between required temporal flexibility and job satisfaction. When OAWT is provided, employees can deal with work issues from anywhere even though the work might not be executed within regular work hours. This aspect is under-emphasized in literature as most studies frame ICTs usage outside work (35) as being predominantly detrimental for employees. We draw attention to the–often neglected–positive role of work connectivity, which helps individuals cope with high demands of required temporal flexibility.

Fourth, by using a sample based on randomization from an exhaustive list of employees in Lower Austria, we distinguish ourselves from commonly used convenience samples. Although a self-selection bias toward higher education is likely, we were able to capture a considerable number of participants that could not access their e-mails outside their office or did not even have professional e-mail accounts reflecting the diversity in the workforce. Furthermore, since the data was collected prior to the pandemic, we can cancel out confounding with the need to work from home and being flexible because of private needs (e.g., homeschooling) or health restrictions (e.g., contagion with the virus).

4.1. Practical implications

The findings of our study have important practical implications related to the debate about blocking OAWT. Although OAWT is directly associated with potentially detrimental aspects of flexibility such as required temporal flexibility, it also relates positively to available temporal and spatial flexibility, which is seen as something beneficial. Results further suggest that having access to work anytime gives employees additional discretionary control over their work and buffers the detrimental impact of required temporal flexibility on job satisfaction. In more detail, our results revealed that when OAWT is high, the relationship between required temporal flexibility and work interrupting nonwork behaviors is positive whereas when OAWT is low, it is negative. Without OAWT, perceived requirements of temporal flexibility might not result in behaviors, but could nonetheless cognitively impact employees' wellbeing (e.g., by thinking on work and knowing that the task cannot be completed).

High work interrupting nonwork behaviors did not relate to low job satisfaction, which is in line with previous research (28, 89). We argue that sometimes it might be better to interrupt nonwork behaviors and finish a work task to satisfy the need for closure (92). This means that blocking employees' e-mail outside work hours may not be the optimal way to deal with employees' connectivity despite the potential downsides of OAWT. Blocking employees' OAWT impairs employees' autonomy (i.e., control). Thus, what seems to be important in organizations is the discussion of required temporal flexibility since the relationship with OAWT is rather high and required temporal flexibility relates negatively with job satisfaction. It is essential that organizations openly communicate their expectations regarding work-extending behaviors and thoroughly reflect their policies about employees' availability outside office hours.

4.2. Limitations

As in every study, this study is subject to several limitations. First, although a representative randomized sample was addressed, there is a self-selection bias. We tried to limit this bias by using paper-pencil questionnaires instead of online questionnaires. Despite this self-selection bias, we were able to use a highly diverse sample to increase generalizability of results.

Second, as in many empirical studies, the results are only based on a cross-sectional sample and no causal conclusions can be made. We tried to collect longitudinal data and asked respondents for their e-mail address for a follow-up study. Only 127 participants provided their e-mail address and were contacted 1.5 years later. Unfortunately, we could only match 25 data sets. Thus, we refrained from testing hypotheses with this limited longitudinal sample.

Third, we obtained a bimodal distribution for OAWT reflecting that many employees either constantly or never have OAWT although normal distribution is preferred in SEM. To deal with non-normality we used MLR estimation in m-plus which is robust to non-normality. Furthermore, the criteria for validity and reliability were also met and thus, we consider the measurement as appropriate. Furthermore, we consider the results about OAWT being the moderating variable as the most important ones, which should not be flawed by the bimodal structure (frequent access vs. no access).

Fourth, aiming to collect a sample that is highly representative, the data involving spatial flexibility is skewed. We had only a low number of employees in our sample who have the possibility for spatial flexibility which might be the reason that the positive relationship of available spatial flexibility on job satisfaction could not reach significance. This might be different nowadays, since many organizations were forced to provide home office to their employees in order to cope with the threats of COVID-19 (93). Thus, our analysis potentially underestimates the positive direct relationship between available spatial flexibility and job satisfaction.

5. Conclusions

Generally, we conclude that OAWT could be seen as a double-edged sword. Our results show that OAWT is associated with available flexibility that relates positively to job satisfaction. However, at the same time, it is also associated with required flexibility, which relates negatively to job satisfaction and positively to work interrupting nonwork behaviors. Our findings also revealed that, OAWT helps strengthen the positive relationship between required temporal flexibility and work interrupting nonwork behaviors. Yet, it also attenuates the negative relationship between required temporal flexibility and job satisfaction. Thus, it is important to critically reflect on the use of ICTs for work outside work hours. Even though OAWT strengthens the relationship between required temporal availability and work interrupting nonwork behaviors, it helps to cope with employees' required flexibility. This particularly needs to be considered when designing flexible working policies in organizations. Despite the general superiority of structural prevention measures over behavioral prevention measures, we highlight that blocking OAWT also reduces boundary control, which in turn, could impair employees' autonomy. Boundary control buffers the negative relationship between ICT demands and rumination of work after work hours (58). Thus, we rather suggest an open discussion about required temporal flexibility in organizations as blocking OAWT would also limit the positive effects of OAWT to cope with required flexibility.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study.

Author contributions

MH-T: planning and execution of the study, data analysis, interpreting results, writing—original draft preparation, and review and editing. AM: interpreting results, writing—original draft preparation, and review and editing. SK: writing—original draft preparation and review and editing. All authors contributed to the article and approved the submitted version.

Acknowledgments

The authors acknowledge TU Wien Bibliothek for financial support through its Open Access Funding Program and the Chamber of Labour of Lower Austria for organizing and financing the distribution of the postal questionnaires to the participants.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

1. Hartner-Tiefenthaler M, Zedlacher E, el Sehity TJ. Remote workers' free associations with working from home during the COVID-19 pandemic in Austria: the interaction between children and gender. Front Psychol. (2022) 13:859020. doi: 10.3389/fpsyg.2022.859020

PubMed Abstract | CrossRef Full Text | Google Scholar

2. Somasundram KG, Hackney A, Yung M, Du B, Oakman J, Nowrouzi-Kia B, et al. Mental and physical health and well-being of Canadian employees who were working from home during the COVID-19 pandemic. BMC Public Health. (2022) 22:1987. doi: 10.1186/s12889-022-14349-5

PubMed Abstract | CrossRef Full Text | Google Scholar

3. Xiao Y, Becerik-Gerber B, Lucas G, Roll SC. Impacts of working from home during COVID-19 pandemic on physical and mental well-being of office workstation users. J Occup Environ Med. (2021) 63:181–90. doi: 10.1097/JOM.0000000000002097

PubMed Abstract | CrossRef Full Text | Google Scholar

4. Ramya SM, Banu J, Asokan Ajitha A, Baral R. Walking on a thin line! Empirical examination of work–home boundary violations faced by employees during forced work from home. Empl Relat. (2022). doi: 10.1108/ER-11-2021-0515. [Epub ahead of print].

CrossRef Full Text | Google Scholar

5. Tedone AM. Keeping up with work email after hours and employee wellbeing: examining relationships during and prior to the COVID-19 pandemic. Occup Health Sci. (2022) 6:51–72. doi: 10.1007/s41542-021-00107-3

PubMed Abstract | CrossRef Full Text | Google Scholar

6. Flecker J, Haidinger B, Schönauer A. Divide and serve: the labour process in service value chains and networks. Competit Change. (2013) 17:6–23. doi: 10.1179/1024529412Z.00000000022

CrossRef Full Text | Google Scholar

7. Messenger JC, Gschwind L. Three generations of telework: New ICTs and the (r)evolution from home office to virtual office. N Technol Work Empl. (2016) 31:195–208. doi: 10.1111/ntwe.12073

CrossRef Full Text | Google Scholar

8. Hassler M, Rau R. Ständige Erreichbarkeit: Flexibilisierungsanforderung oder Flexibilisierungsmöglichkeit. Wirtschaftspsychologie. (2016) 2:25–34.

Google Scholar

9. Menz W, Pauls N, Pangert B. Arbeitsbezogene erweiterte Erreichbarkeit: Ursachen, Umgangsstrategien und Bewertung am Beispiel von IT Beschaeftigten. Wirtschaftspsychologie. (2016) 18:55–66.

Google Scholar

10. Chung H, van der Horst M. Women's employment patterns after childbirth and the perceived access to and use of flexitime and teleworking. Hum Relat. (2018) 71:47–72. doi: 10.1177/0018726717713828

PubMed Abstract | CrossRef Full Text | Google Scholar

11. Kelly E L, Moen P, Tranby E. Changing workplaces to reduce work-family conflict: Schedule control in a white-collar organization. Am Sociol Rev. (2011) 76:265–90. doi: 10.1177/0003122411400056

PubMed Abstract | CrossRef Full Text | Google Scholar

12. Bitkom. Work 3.0 - Working in the Digital World. (2013). Available online at: https://www.bitkom.org/sites/main/files/file/import/Studie-Arbeit-30.pdf (accessed January 10, 2023).

Google Scholar

13. Schöllbauer J, Hartner-Tiefenthaler M, Kelliher C. ICT-enabled work extension and its consequences: A paradoxical situation between high performance and low wellbeing. In:Korunka C, , editor. Flexible Working Practices and Approaches: Psychological and Social Implications of a Multifaceted Phenomenon. Cham: Springer Nature Switzerland AG. (2021). p. 49–166.

Google Scholar

14. Kossek EE, Ruderman MN, Braddy PW, Hannum KM. Work–nonwork boundary management profiles: a person-centered approach. J Vocat Behav. (2012) 81:112–28. doi: 10.1016/j.jvb.2012.04.003

CrossRef Full Text

15. Capitano J, DiRenzo MS, Aten KJ, Greenhaus JH. Role identity salience and boundary permeability preferences: an examination of enactment and protection effects. J Vocat Behav. (2017) 102:99–111. doi: 10.1016/j.jvb.2017.07.001

CrossRef Full Text | Google Scholar

16. Kossek EE, Lautsch BA. Work–family boundary management styles in organizations. Organ Psychol Rev. (2012) 2:152–71. doi: 10.1177/2041386611436264

CrossRef Full Text | Google Scholar

17. Allvin M, Mellner C, Movitz, Aronsson G. The diffusion of flexibility: Estimating the incidence of low-regulated working conditions. Nord J Work Life Stud. (2013) 3:99. doi: 10.19154/njwls.v3i3.3013

CrossRef Full Text | Google Scholar

18. Wessels C, Schippers MC, Stegmann S, Bakker AB, van Baalen PJ, Proper KI. Fostering flexibility in the new world of work: a model of time-spatial job crafting. Front Psychol. (2019) 10. doi: 10.3389/fpsyg.2019.00505

PubMed Abstract | CrossRef Full Text | Google Scholar

19. Cooper CL, Lu L. Excessive availability for work: Good or bad? Charting underlying motivations and searching for game-changers. Hum Resour Manag Rev. (2019) 29:100682. doi: 10.1016/j.hrmr.2019.01.003

CrossRef Full Text | Google Scholar

20. Kerman K, Tement S, Korunka C. Don't leave your heart at work: Profiles of work–life interference and cardiometabolic risk. Int J Stress Manag. (2021) 28:105–16. doi: 10.1037/str0000224

CrossRef Full Text | Google Scholar

21. Schlachter S, McDowall A, Cropley M, Inceoglu I. Voluntary work-related technology use during non-work time: a narrative synthesis of empirical research and research agenda. Int J Manag Rev. (2018) 20:825–46. doi: 10.1111/ijmr.12165

CrossRef Full Text | Google Scholar

22. European Commission Council Council Directive of 12 June 1989 on the introduction of measures to encourage improvements in the safety and health of workers at work (89/391/EEC). Available online at: https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A01989L0391-20081211 (accessed January 10, 2023).

23. Kaufmann M,. Erreichbarkeit nach Dienstschluss: Deutsche Konzerne kämpfen gegen den Handy-Wahn. KarriereSpiegel. (2014). Available online at: http://www.spiegel.de/karriere/erreichbar-nach-dienstschluss-massnahmen-der-konzerne-a-954029.html (accessed January 12, 2018).

24. Nerdinger FW, Blickle G, Schaper N. Arbeits- und Organisationspsychologie (4th ed). Berlin: Springer. (2019).

Google Scholar

25. Saari LM, Judge TA. Employee attitudes and job satisfaction. Hum Resour Manage. (2004) 43:395–407. doi: 10.1002/hrm.20032

CrossRef Full Text | Google Scholar

26. Faragher EB, Cass M, Cooper CL. The relationship between job satisfaction and health: a meta-analysis. Occup Environ Med. (2005) 62:105–12. doi: 10.1136/oem.2002.006734

PubMed Abstract | CrossRef Full Text | Google Scholar

27. Bal PM, Izak M. Paradigms of flexibility: a systematic review of research on workplace flexibility. Eur Manag Rev. (2021) 18:37–50. doi: 10.1111/emre.12423

CrossRef Full Text | Google Scholar

28. Diaz I, Chiaburu DS, Zimmerman RD, Boswell WR. Communication technology: pros and cons of constant connection to work. J Vocat Behav. (2012) 80:500–8. doi: 10.1016/j.jvb.2011.08.007

CrossRef Full Text | Google Scholar

29. Kelliher C, Richardson J, Boiarintseva G. All of work? All of life? Reconceptualising work-life balance for the 21st century. Hum Resour Manag J. (2019) 29:97–112. doi: 10.1111/1748-8583.12215

CrossRef Full Text | Google Scholar

30. Leung L, Zhang R. Mapping ICT use at home and telecommuting practices: a perspective from work/family border theory. Telemat Inform. (2017) 34:385–96. doi: 10.1016/j.tele.2016.06.001

CrossRef Full Text | Google Scholar

31. Raghuram S, Hill NS, Gibbs JL, Maruping LM. Virtual work: bridging research clusters. Acad Manag Ann. (2019) 13:308–41. doi: 10.5465/annals.2017.0020

CrossRef Full Text | Google Scholar

32. Spieler I, Scheibe S, Stamov-Roßnagel C, Kappas A. Help or hindrance? Day-level relationships between flextime use, work-nonwork boundaries, and affective wellbeing. J Appl Psychol. (2017) 102:67–87. doi: 10.1037/apl0000153

PubMed Abstract | CrossRef Full Text | Google Scholar

33. Der Spiegel,. Porsche-Betriebsratschef will E-Mails löschen lassen. (2017). Available online at: https://www.spiegel.de/karriere/porschebetriebsratschef-will-e-mails-nach-feierabend-loeschen-lassen-a-1183842.html (accessed May 17, 2022).

Google Scholar

34. Adisa TA, Gbadamosi G, Osabutey ELC. What happened to the border? The role of mobile information technology devices on employees' work-life balance. Pers Rev. (2017) 46:1651–71. doi: 10.1108/PR-08-2016-0222

CrossRef Full Text | Google Scholar

35. Dettmers J, Vahle-Hinz T, Bamberg E, Friedrich N, Kelle M. Extended work availability and its relation with start-of-day mood and cortisol. J Occup Health Psychol. (2016) 21:105–18. doi: 10.1037/a0039602

PubMed Abstract | CrossRef Full Text | Google Scholar

36. Porter S, Ayman R. Work flexibility as a mediator of the relationship between work-family conflict and intention to quit. J Manag Organ. (2010) 16:411–24. doi: 10.5172/jmo.16.3.411

CrossRef Full Text | Google Scholar

37. Mustafa M, Gold M. ‘Chained to my work'? Strategies to manage temporal and physical boundaries among self-employed teleworkers. Hum Resour Manag J. (2013) 23:413–29. doi: 10.1111/1748-8583.12009

CrossRef Full Text | Google Scholar

38. Allen TD, Johnson RC, Kiburz KM, Shockley KM. Work-Family conflict and flexible working arrangements: Deconstructing flexibility. Pers Psychol. (2013) 66:345–76. doi: 10.1111/peps.12012

CrossRef Full Text | Google Scholar

39. Galinsky E, Bond, JT, Sakai, K,. National Study of Employers. (2008). Available online at: http://familiesandwork.org/site/research/reports/2008nse.pdf (accessed January 12, 2016).

Google Scholar

40. Gajendran RS, Harrison DA. The good, the bad, and the unknown about telecommuting: meta-analysis of psychological mediators and individual consequences. J Appl Psychol. (2007) 92:1524–41. doi: 10.1037/0021-9010.92.6.1524

PubMed Abstract | CrossRef Full Text | Google Scholar

41. McNall LA, Masuda AD, Nicklin JM. Flexible working arrangements, job satisfaction, and turnover intentions: the mediating role of work-to-family enrichment. J Psychol. (2010) 144:61–81. doi: 10.1080/00223980903356073

PubMed Abstract | CrossRef Full Text | Google Scholar

42. De Menezes LM, Kelliher C. Flexible working, individual performance, and employee attitudes: comparing formal and informal arrangements. Hum Resour Manage. (2017) 56:1051–70. doi: 10.1002/hrm.21822

CrossRef Full Text | Google Scholar

43. Bakker AB, Demerouti E. The job demands-resources model: State of the art. J Manag Psychol. (2007) 22:309–28. doi: 10.1108/02683940710733115

PubMed Abstract | CrossRef Full Text | Google Scholar

44. Chung-Yan GA. The nonlinear effects of job complexity and autonomy on job satisfaction, turnover, and psychological wellbeing. J Occup Health Psychol. (2010) 15:237–51. doi: 10.1037/a0019823

PubMed Abstract | CrossRef Full Text | Google Scholar

45. Goñi-Legaz S, Ollo-López A. Temporary contracts, participation in decision making and job satisfaction in European workers. Int J Manpow. (2017) 38:875–92. doi: 10.1108/IJM-04-2016-0086

CrossRef Full Text | Google Scholar

46. Hackman JR, Oldham GR. Development of the job-diagnostic-survey. J Appl Psychol. (1975) 60:159–70. doi: 10.1037/h0076546

CrossRef Full Text | Google Scholar

47. Häusser JA, Mojzisch A, Schulz-Hardt S. Ten years on: a review of recent research on the job demand-control (-support) model and psychological well-being. Work Stress. (2010) 24:1–35. doi: 10.1080/02678371003683747

PubMed Abstract | CrossRef Full Text | Google Scholar

48. Spector PE. Perceived control by employees: a meta-analysis of studies concerning autonomy and participation at work. Hum Relat. (1986) 39:1005–16. doi: 10.1177/001872678603901104

CrossRef Full Text | Google Scholar

49. Jarvenpaa SL, Lang KR. Managing the paradoxes of mobile technology. Inf Syst Manag. (2005) 22:7–23. doi: 10.1201/1078.10580530/45520.22.4.20050901/90026.2

CrossRef Full Text | Google Scholar

50. Mazmanian M, Orlikowski WJ, Yates J. The autonomy paradox: the implications of mobile email devices for knowledge professionals. Organ Sci. (2013) 24:1337–57. doi: 10.1287/orsc.1120.0806

CrossRef Full Text | Google Scholar

51. Pongratz HJ, Voß GG. From employee to “entreployee”: Towards a “self-entrepreneurial” work force? Concepts Transform. (2003) 8:239–54. doi: 10.1075/cat.8.3.04pon

PubMed Abstract | CrossRef Full Text | Google Scholar

52. Day A, Paquet S, Scott N, Hambley L. Perceived information and communication technology (ICT) demands on employee outcomes: The moderating effect of organizational ICT support. J Occup Health Psychol. (2012) 17:473–91. doi: 10.1037/a0029837

PubMed Abstract | CrossRef Full Text | Google Scholar

53. Höge T. Perceived flexibility requirements at work and the entreployee-work-orientation: concept and measurement. Psychol Everyday Act. (2011) 4:3–21. Available online at: http://www.allgemeine-psychologie.info/cms/images/stories/allgpsy_journal/Vol%204%20No%201/hoege.pdf

Google Scholar

54. Höge T, Hornung S. Perceived flexibility requirements: Exploring mediating mechanisms in positive and negative effects on worker well-being. Econ Ind Democr. (2015) 36:407–30. doi: 10.1177/0143831X13511274

CrossRef Full Text | Google Scholar

55. Kattenbach R, Demerouti E, Nachreiner F. Flexible working times: Effects on employees' exhaustion, work-nonwork conflict and job performance. Career Dev Int. (2010) 15:279–95. doi: 10.1108/13620431011053749

CrossRef Full Text | Google Scholar

56. De Castella T. Could work emails be banned after 6pm?. BBC News Magazine. (2014). Available online at: http://www.bbc.com/news/magazine-26958079 (accessed January 12, 2016).

Google Scholar

57. Ashforth BE, Kreiner GE, Fugate M. All in a day's work: boundaries and micro role transitions. Acad Manag Rev. (2000) 25:472–91. doi: 10.2307/259305

CrossRef Full Text | Google Scholar

58. Park Y, LiuY, Headrick L. When work is wanted after hours: Testing weekly stress of information communication technology demands using boundary theory. J Org Behav. (2020) 41:518–34. doi: 10.1002/job.2461

CrossRef Full Text | Google Scholar

59. Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. The job demands-resources model of burnout. J Appl Psychol. (2001) 86:499–512. doi: 10.1037/0021-9010.86.3.499

CrossRef Full Text | Google Scholar

60. Schaufeli WB, Bakker AB. Job demands, job resources, and their relationship with burnout and engagement: a multi-sample study. J Organ Behav. (2004) 25:293–315. doi: 10.1002/job.248

CrossRef Full Text | Google Scholar

61. Greenhaus JH, Beutell NJ. Sources of conflict between work and family roles. Acad Manag Rev. (1985) 10:76. doi: 10.2307/258214

CrossRef Full Text | Google Scholar

62. Olson-Buchanan JB, Boswell WR. Blurring boundaries: Correlates of integration and segmentation between work and nonwork. J Vocat Behav. (2006) 68:432–45. doi: 10.1016/j.jvb.2005.10.006

CrossRef Full Text | Google Scholar

63. AK Niederösterreich. Work Climate in Flexible Teams. Unpublished report: Chamber of Labour of Lower Austria. (2019).

PubMed Abstract

64. Søgaard AJ, Selmer R, Bjertness E, Thelle D. The oslo health study: the impact of self-selection in a large, population-based survey. Int J Equity Health. (2004) 3:3. doi: 10.1186/1475-9276-3-3

PubMed Abstract | CrossRef Full Text | Google Scholar

65. Muse LA, Pichler S. A comparison of types of support for lower-skill workers: Evidence for the importance of family supportive supervisors. J Vocat Behav. (2011) 79:653–66. doi: 10.1016/j.jvb.2011.04.005

CrossRef Full Text | Google Scholar

66. Gerdenitsch C, Dorn R, Prem, R, Korunka C. Flexible working – Autonomy and restriction? In: Paper presented at the 11th Conference of the Austrian Society of Psychology. Vienna. (2014).

PubMed Abstract | Google Scholar

67. Brislin RW. Back-translation for cross-cultural research. J Cross Cult Psychol. (1970) 1:185–216. doi: 10.1177/135910457000100301

CrossRef Full Text

68. Gittell JH, Weinberg D, Pfefferle S, Bishop C. Impact of relational coordination on job satisfaction and quality outcomes: a study of nursing homes. Hum Resour Manag J. (2008) 18:154–70. doi: 10.1111/j.1748-8583.2007.00063.x

CrossRef Full Text | Google Scholar

69. Nagy MS. Using a single-item approach to measure facet job satisfaction. J Occup Organ Psychol. (2002) 75:77–86. doi: 10.1348/096317902167658

CrossRef Full Text | Google Scholar

70. Scarpello V, Campbell JP. Job satisfaction: Are all the parts there? Pers Psychol. (1983) 36:577–600. doi: 10.1111/j.1744-6570.1983.tb02236.x

CrossRef Full Text | Google Scholar

71. Wohlers C, Hartner-Tiefenthaler M, Hertel G. The relation between activity-based work environments and office workers' job attitudes and vitality. Environ Behav. (2019) 51:167–98. doi: 10.1177/0013916517738078

CrossRef Full Text | Google Scholar

72. Wanous JP, Hudy MJ. Single-item reliability: a replication and extension. Organ Res Methods. (2001) 4:361–75. doi: 10.1177/109442810144003

CrossRef Full Text | Google Scholar

73. Wanous JP, Reichers AE, Hudy MJ. Overall job satisfaction: how good are single-item measures? J Appl Psychol. (1997) 82:247–52. doi: 10.1037/0021-9010.82.2.247

PubMed Abstract | CrossRef Full Text | Google Scholar

74. Eikhof DR, Warhurst C, Haunschild A. Introduction: what work? What life? What balance? Empl Relat. (2007) 29:325–33. doi: 10.1108/01425450710839452

PubMed Abstract | CrossRef Full Text | Google Scholar

75. Liu H, Wang Q, Keesler V, Schneider B. Non-standard work schedules, work–family conflict and parental well-being: a comparison of married and cohabiting unions. Soc Sci Res. (2011) 40:473–84. doi: 10.1016/j.ssresearch.2010.10.008

CrossRef Full Text | Google Scholar

76. Judge TA, Watanabe S. (1993). Another look at the job satisfaction-life satisfaction relationship. J Appl Psychol. (2011) 78:939–48. doi: 10.1037/0021-9010.78.6.939

CrossRef Full Text | Google Scholar

77. Cheung F, Lucas RE. Assessing the validity of single-item life satisfaction measures: results from three large samples. Qual Life Res. (2014) 23:2809–18. doi: 10.1007/s11136-014-0726-4

PubMed Abstract | CrossRef Full Text | Google Scholar

78. Anderson JC, Gerbing DW. Structural equation modeling in practice: a review and recommended two-step approach. Psychol Bull. (1988) 103:411–23. doi: 10.1037/0033-2909.103.3.411

CrossRef Full Text | Google Scholar

79. Muthén LK, Muthén BO. Mplus User's Guide. Eighth Edition (1998-2017). Los Angeles, CA: Muthén & Muthén.

Google Scholar

80. Williams LJ, Vandenberg RJ, Edwards JR. Structural equation modeling in management research: a guide for improved analysis. Acad Manag Ann. (2009) 3:543–604. doi: 10.5465/19416520903065683

CrossRef Full Text | Google Scholar

81. Little TD, Bovaird JA, Widaman KF. On the merits of orthogonalizing powered and product terms: Implications for modeling interactions among latent variables. Struct Equ Model. (2006) 13:497–519. doi: 10.1207/s15328007sem1304_1

CrossRef Full Text | Google Scholar

82. Steinmetz H, Davidov E, Schmidt P. Three approaches to estimate latent interaction effects: Intention and perceived behavioral control in the theory of planned behavior. Methodol Innov. (2011) 6:95–110. doi: 10.4256/mio.2010.0030

CrossRef Full Text | Google Scholar

83. Koufteros XA. Testing a model of pull production: a paradigm for manufacturing research using structural equation modeling. J Oper Manag. (1999) 17:467–488. doi: 10.1016/S0272-6963(99)00002-9

CrossRef Full Text | Google Scholar

84. Bagozzi RP, Yi Y. On the evaluation of structural equation models. J Acad Mark Sci. (1988) 16:74–94. doi: 10.1007/BF02723327

CrossRef Full Text | Google Scholar

85. Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. J Mark Res. (1981) 18:39–50. doi: 10.1177/002224378101800104

CrossRef Full Text | Google Scholar

86. Liang H, Saraf N, Hu Q, Xue Y. Assimilation of enterprise systems: the effect of institutional pressures and the mediating role of top management. MIS Q. (2007) 31:59. doi: 10.2307/25148781

CrossRef Full Text | Google Scholar

87. Becker W. Employer expectations on off-hours email: New study shows adverse health effects on employees and families. Virginia Tech Daily. (2018). Available online at: https://vtnews.vt.edu/articles/2018/08/pamplin-employer-emailexpectations.html (accessed July 10, 2019).

88. Adisa T, Ogbonnaya C, Adekoya OD. Remote working and employee engagement: a qualitative study of British workers during the pandemic. Inf Technol People. (2021). doi: 10.1108/ITP-12-2020-0850. [Epub ahead of print].

CrossRef Full Text | Google Scholar

89. Moore AT. Computers, cell phones, and social media: how after-hours communication impacts work–life balance and job satisfaction (dissertation). Virginia Beach, Regent University, Virginia, VA, United States. (2017).

90. Derks D, Bakker AB. Smartphone use, work-home interruption, and burnout: a diary study on the role of recovery. Appl Psychol. (2014) 63:411–40. doi: 10.1111/j.1464-0597.2012.00530.x

CrossRef Full Text | Google Scholar

91. Kinnunen U, Feldt T, Bloom J, de Sianoja M, Korpela K, Geurts S. Linking boundary crossing from work to nonwork to work-related rumination across time: a variable- and person-oriented approach. J Occup Health Psychol. (2017) 22:467–80. doi: 10.1037/ocp0000037

PubMed Abstract | CrossRef Full Text | Google Scholar

92. Weigelt O, Syrek CJ. Ovsiankina's great relief: How supplemental work during the weekend may contribute to recovery in the face of unfinished tasks. Int J Environ Res Publ Health. (2017) 14:1606. doi: 10.3390/ijerph14121606

PubMed Abstract | CrossRef Full Text | Google Scholar

93. Desilver D,. Before the Coronavirus, Telework was an Optional Benefit, Mostly for the Affluent Few. Pew Research Center. (2020). Available online at: https://www.pewresearch.org/fact-tank/2020/03/20/before-the-coronavirus-telework-was-an-optional-benefit-mostly-for-the-affluent-few/ (accessed March 30, 2020).