The Hidden Positive Side of Constant Connectivity
How today’s work reality facilitates employee engagement
Kathrin Fischer 12530298
Graduate School of Communication, University of Amsterdam Master’s Programme Communication Science
Master’s Thesis Dhr. Dr. Ward van Zoonen
Abstract
Mobile work devices created a context of constant connectivity that technically enables employees to connect to work anywhere, anytime. To date, investigations on employees’ coping strategies have either focused on demanding aspects of connectivity with detrimental outcomes for psychological engagement; demonstrated enhanced engagement in form of behavioural activities; or considered engagement as an antecedent that precedes desirable work outcomes. What is missing, is quantitative longitudinal research that provides support for causal effects and considers engagement as a positive mindset. This is crucial because connectivity is here to stay and requires organisational policies to be designed in a way that encourages favourable engagement and does not impede it. Hence, this study investigates the extent to which constant connectivity during non-worktime relates to engagement understood as a positive mindset, including heightened vigour, dedication and absorption. Results from two survey waves (N = 345) at a corporation support the hypothesized causal effects and demonstrate that constant connectivity facilitates employees’ engagement through leveraging their autonomy. This embodies two contributions. First, constant connectivity has causal priority over engagement and is thus a functional resource that interacts with other job resources. Second, its interaction with autonomy is crucial because it makes the paramount relationship between connectivity and engagement detectable.
Keywords: constant connectivity, employee engagement, autonomy, job resources, mobile work devices.
Introduction
The influx of mobile work devices (MWDs) such as laptops and smartphones has led to radical transformations of work, since it created a context of potential constant connectivity in many contemporary workplaces (Kolb, Caza, & Collins, 2012; Mazmanian, 2013).
Constant connectivity is defined as the state of being available and responsive regardless of time and place (Wajcman & Rose, 2011), which often embodies a perception of being constantly engaged with work (MacCormick, Dery, & Kolb, 2012), also during after-hours and weekends (Mazmanian, Orlikowski, & Yates, 2013). Concomitantly, interest in unveiling employees’ coping strategies with connectivity during non-worktime and implications for their engagement has grown (Bakker, Albrecht, & Leiter, 2011; Büchler, ter Hoeven, & van Zoonen, 2020; Derks & Bakker, 2010), and lies at the core of this study.
Employee engagement, if present, can play a powerful role in creating distinct
competitive advantage for organisations because it incorporates an organisational purpose and favourable attitudinal and behavioural components (Macey & Schneider, 2008). To date, the concept is ambiguously used to refer to psychological states (e.g. vigour, dedication,
absorption), traits (e.g. positive affect), and behaviours (e.g. observable performance), and to interpret antecedents and outcomes of these (Macey & Schneider, 2008). Studies on constant connectivity during non-worktime arrive at mixed interpretations around its potential in facilitating or impeding engagement, and conceptualize engagement inconsistently.
The figure-ground perspective views connectivity as a demanding aspect (Bordi, Okkonen, Mäkiniemi, & Heikkilä-Tammi, 2018) and impediment for positive psychological engagement (Boswell & Olson-Buchanan, 2007; Büchler et al., 2020). It demonstrates that MWDs enable employees to connect to work anywhere, anytime, so that work creeps into times and places which were previously reserved for family and leisure (Chesley, 2005). As such, constant connectivity impedes the process of mental detachment from work (ter Hoeven
& van Zoonen, 2020) with detrimental outcomes for employees, such as poor wellbeing (Büchler et al., 2020), work-life conflict (Boswell & Olson-Buchanan, 2007), and escalating engagement characterized by working everywhere, all the time (Mazmanian et al., 2013).
Notwithstanding these downsides, a second string of research elucidates that constant connectivity is not a matter of choice anymore – it is required for organisations to thrive and for professionals to work (Barley, Meyerson, & Grodal, 2011; MacCormick et al., 2012; Kolb et al., 2012; Wajcman & Rose, 2011). Particularly for knowledge professionals, connectivity is regarded indispensable because they fully rely on information sharing to accomplish their tasks (Waizenegger, Ulrich, & Maier, 2014). While this string of research also departs from the point that constant connectivity enables employees to stay on top of things anywhere, anytime (Kossek & Lautsch, 2012; Kreiner, Hollensbe & Sheep, 2009), it concludes more positively and highlights outcomes such as enhanced productivity (Mazmanian et al., 2013), as well as increased ability to perform work and opportunities to engage (Waizenegger, Ulrich, & Maier, 2014). Hence, it establishes the assumption that constant connectivity can and does facilitate engagement.
However, in this string of research engagement is conceptualized as an observable behaviour showing that individuals engage with MWDs, signal their availability and put forth discretionary effort in the form of extra time and energy (Towers-Perrin, 2003). Yet, defining engagement solely in terms of doing more than what is usual, is disadvantageous, because psychologically engaged employees were shown to not only do more, but to do something different (Brown, 1996; Kahn, 1990). They include a sense of self-identity with the work they do, draw more on their selves in their roles and thus, perform more stirringly (Kahn, 1990). Hence, there exists an important distinction between the experiential state (i.e. psychological engagement) and observable behaviours (i.e. performance) that may or may not accompany that state (Kahn, 1992). This separate focus is crucial to distinguish between psychological
outcomes that are personally relevant and those that are organisationally relevant (Macey & Schneider, 2008). Beneficial outcomes for organisations typically emerge when engagement behaviours are accompanied by favourable states. Yet, this potential remains unexplored, since engagement as a positive psychological state is largely disregarded by connectivity literature.
Thus, it remains puzzling whether positive psychological engagement accompanies such observable behaviours and whether constant connectivity during non-worktime is helpful in creating it, or only impedes it as the figure-ground perspective claims. For most scholars the attributes of tasks are key to enhance engagement. As such, they refer to job characteristics (Hackman, 1980) and studies on the intrinsic nature of rewards (Gagne & Deci, 2005) to specify issues that drive employees’ passion, commitment, and involvement among other indicators. Although the task is central, it is the degree to which employees can implement their preferred self in the work that seems to matter most for leveraged
engagement (Macey & Schneider, 2008). This directly points towards the need to consider more thoroughly, not only the conditions surrounding workplaces, but also to examine the underlying processes through which individuals internalize external factors.
To date, it remains unclear whether constant connectivity is a behavioural outcome that is observable as a result of an employee’s psychological state, or whether it is a feature of contemporary workplaces that enables employees to become more psychologically engaged. This unclarity might ground in the inconsistent application of engagement as a concept, as well as the absence of quantitative longitudinal research, thereby providing limited support for causal effects. However, clarifying these effects is fundamental because it advances our understanding of what connectivity behaviours are, and how organisations should view and address them. Therefore, this study examines the research question: to what extent does constant connectivity during non-worktime relate to favourable psychological engagement.
Thereby, it seeks to make two contributions. First, it aims to establish the extent to which constant connectivity benefits professionals in terms of facilitating engagement. If it indeed enables employees to become more engaged, it could be considered a valuable resource in today’s workplaces that is important in its own right (Hobfoll, 2002). This plays into the second intended contribution: understanding the role of autonomy in this relationship. If constant connectivity supports autonomy, it helps to explain the causal direction between constant connectivity and engagement. More leeway for instance, in how, where, and when to work, makes the causal effect from constant connectivity to engagement more likely because employees are technically enabled to harness more of their individual self with their work (Kahn, 1990), which ultimately speaks for enhanced psychological engagement.
Theoretical background
To understand the relationship between constant connectivity during non-worktime and engagement, this study departs from literature that reveals various conceptualizations of engagement and stresses the importance of work conditions in facilitating it (Bakker et al., 2011; Macey & Schneider, 2008). Considering constant connectivity an inherent condition of today’s workplaces (Kolb et al., 2012), it links these notions with empirical investigations on constant connectivity and presents five anticipated effects.
Four direct effects
Constant connectivity and engagement
Constant connectivity during non-worktime is established as availability and connectedness, checking and answering of messages, and basis for employees to conduct work remotely (Büchler et al., 2020). In contrast, engagement is regarded a “concept with a sparse and diverse theoretical and empirically demonstrated nomological net” (Macey & Schneider, 2008, p. 3) because relationships among potential antecedents and consequences of engagement, as well as its components are conceptualized and interpreted very inconsistently.
Predominantly, engagement is conceptualized as an employee trait (i.e. a stable characteristic including positive affect), a behaviour (i.e. an adaptive performance), or a psychological state (i.e. a cognitive condition including heightened energy, vigour, effort, dedication, absorption, commitment, resilience) (Macey & Schneider, 2008). Here, engagement is understood as a persistent affective-motivational state of fulfilment in employees that is characterized by vigour, dedication, and absorption (Schaufeli, Salanova, Gonzalez-Roma, & Bakker, 2002). Thus, it captures a positive state that employees
experience through their work: as something stimulating and energetic to which they want to devote time, effort and persistence (vigour); as a significant and meaningful pursuit in which they are willing to invest themselves (dedication); and something on which they are fully concentrated and have difficulties in detaching themselves (absorption) (Bakker et al., 2011). Since scholars have positioned this form of engagement as an antecedent for behaviours that are thought to be of value to organisational effectiveness, as well as an attitudinal outcome of external work conditions (Macey & Schneider, 2008), this study departs from two
overarching assumptions.
Engagement as an antecedent. The prevailing assumption in literature is that trait
and/or state engagement precede desirable work behaviours, so that engagement enhances positive employee and work outcomes (Bakker et al., 2011). When viewing engagement as a pre-existing cognitive state or trait, communication is predominantly conceptualized as a discretionary effort or behaviour of employees (Macey & Schneider, 2008). Communication is used to simply express or demonstrate employees’ pre-existing positive affect toward work.
Moreover, using MWDs to connect and communicate during non-worktime is not something that just ‘happens’ to employees but employees are rather agentic in managing their connectivity behaviours (Russo, Ollier-Malaterre, & Morandin, 2019). Depending on the centrality of work in their lives, they construct their own use patterns of MWDs (Dery &
MacCormick, 2012) and conserve their engagement through a process of job crafting to maintain or increase it (Bakker et al., 2011; Wrzesniewski & Dutton, 2001). Employees who are fulfilled by their work and find it enjoyable, are putting forth discretionary effort, extra time, energy, persistence and initiative (Towers-Perrin, 2003). Hence, among these other indicators, they are likely to invest more time (Rothbard & Edwards, 2003) because their job role is consistent with their personal goals and important to them in terms of their self-identity. Also, they face difficulties with detaching themselves from work (Schaufeli & Bakker, 2004). Applying this prevailing notion to the context at hand, it is first tested whether engaged employees are more connected during non-worktime. The departing suggestion is:
Hypothesis 1a: Engagement positively affects constant connectivity.
Engagement as an outcome. Yet, it is limiting to understand engagement solely in
terms of extra effort and doing more than what is usual (Kahn, 1992). Engagement does not “spring forth whole”, but is dependent on the conditions under which employees work (Macey & Schneider, 2008, p. 19). As an inherent characteristic of today’s workplaces (Kolb et al., 2012), constant connectivity embodies such a potential antecedent that can enhance engagement as an outcome. It is this perspective, on which this study focuses.
Scholars outline that constant connectivity provides employees with a sense of being in the loop, and thereby allows them better task fulfilment, enhanced performance and engagement (Cavazotte et al., 2014; Kolb et al., 2012; MacCormick et al., 2012). Thus, in connectivity literature, the prevailing notion is that constant connectivity is an antecedent for (dis)engagement. When investigated as a positive outcome, engagement is not conceptualized as a psychological state but as an activity, i.e. humans enacting connectivity or using MWDs to communicate. In consequence, engagement actually measures the same as connectivity, so that the often-suggested positive relationship appears meaningless. However, investigating psychological engagement “additionally refers to the investment of the self in the person’s
work and the perceived importance of work outcomes … to that person’s identity” (Macey & Schneider, 2008, p. 13). Hence, this study makes an important contribution to the
connectivity literature and answers a call to investigate how and why certain tactics are used to facilitate engagement (Bakker et al., 2011).
Psychological engagement goes beyond simple task fulfilment and implies that
employees draw on their selves in their roles, and are attentive, focused and absorbed (Macey & Schneider, 2008). Van Zoonen and Banghart (2018) reveal that employees rather construct and enact engagement through communicating for and about their work, instead of merely demonstrating it in communication. While van Zoonen and Banghart (2018) focus on work-related social media use, this lens can also be applied to constant connectivity. Following the constructivist communication perspective (Fairhurst & Putnam, 2004; 2014), phenomena such as connectivity behaviours are not merely products of pre-existing states (e.g. engagement) but also embody processes through which engagement is created.
As employees actively craft their jobs, they draw upon their skills, abilities and resources to minimize job demands (Bakker et al., 2011). Constant connectivity during non-worktime might be one of those resources or communicative behaviours to address work demands and facilitate a favourable psychological state. For instance, it enables employees to manage work on their own terms since it allows them to be on top of things and come to the office prepared (Cavazotte et al., 2014; Gruber, Sarigianni, Geiger, & Remus, 2018; Kossek & Lautsch, 2012). Reading emails during commute allows employees to handle “more trivial tasks” outside workhours, which yields more time at the office to concentrate on
“intellectually demanding tasks, or real work” (Cavazotte et al., 2014, p. 83). This is often experienced as a feeling of relief and reduced volume of work, through which professionals demonstrate more enthusiasm, and a sense of being better able to exercise their competences.
Accordingly, the prevailing assumption in literature is reversed here, and it is assumed that communicative behaviours emerging from the context of constant connectivity can play a fundamental role in constructing psychological engagement. This leads to:
Hypothesis 1b: Constant connectivity positively affects engagement.
Constant connectivity and autonomy
As constant connectivity has become an inherent characteristic of today’s workplaces, employees are technically enabled to work anywhere, anytime (Perlow, 2012). This relates directly to the concept of autonomy at the workplace, which is defined as “the degree to which the job provides substantial freedom, independence, and discretion to the individual in scheduling the work and determining the procedures to … carrying it out” (Hackman & Oldham, 1975, p. 162). Literature to date has either established a positive relationship
between the constructs, or conceptualized it as a double-edged sword, and thereby questioned the very nature of this relationship (Cavazotte et al., 2014; Mazmanian et al., 2013).
Mazmanian et al. (2013) establish that connectivity during non-worktime enables professionals to continually monitor the flow of communication, while choosing whether and when to participate. This is experienced as greater discretion, flexibility and control.
However, through qualitative interviews they also find that the intensified communication practice of individuals concomitantly shifts collective norms towards escalating expectations of accessibility and responsiveness, thereby encroaching upon employees’ autonomy
(Mazmanian et al., 2013). Cavazotte et al. (2014) highlight that the ability to be more flexible and quicker in the management of tasks increases employees’ perceived autonomy, while they are equally aware of the fact that work inescapably creeps into unconventional time slots and private occasions, thereby making it difficult to detach. Consequently, connectivity is not only experienced as autonomy enhancing, but simultaneously as autonomy limiting (i.e. as a
paradox). Yet, to date, quantitative longitudinal research designs are missing, providing limited support for such conceptualized ambiguous relationships.
Other studies suggest a solely positive relationship between constant connectivity and autonomy (Chesley, 2006; Gruber et al., 2018; Kelliher & Anderson, 2010; Kossek, Lautsch, & Eaton, 2006; Piszczek, 2017; ter Hoeven & van Zoonen, 2020). Also, this study assumes a positive relationship based on the self-determination theory that indicates that humans are intrinsically motivated to fulfil their psychological needs (Ryan & Deci, 2000). One of these psychological needs is autonomy, again implying “the need to act with a sense of choice and volition” (ter Hoeven & van Zoonen, 2020, p. 15).
The fulfilment of basic needs can either be supported or hindered by external contexts (Ryan & Deci, 2000). Empirical research on constant connectivity and autonomy (regardless of establishing a reciprocal or positive relationship) shows that connectivity can support the fulfilment of needs. For instance, it “enables work practices to be redesigned in a way that allows individuals to connect with people, information and activities anytime, anywhere” (Dery & MacCormick, 2012, p. 168). Thereby, it allows employees to alter their work
procedures and have freedom in how to design this change. Also, it embodies the opportunity to transcend temporal and spatial boundaries at employees’ discretion, and thereby enables them to decide more freely when and where to work (Gruber et al., 2018). This is not only experienced as greater flexibility, but also as greater control over different life aspects, such as work and family (Kossek et al., 2006).
The above suggests that autonomy at the workplace is experienced in various ways, most prominently, in the freedom to decide how, where and when to work. Breaugh (1985) established three areas (facets) of autonomy, which this study follows. Work method autonomy refers to “the degree of discretion … individuals have regarding the procedures they utilize in going about their work” (Breaugh, 1985, p. 556). Work scheduling autonomy
implies “the extent to which employees feel they can control the scheduling/ sequencing/ timing of their work”, and work criteria autonomy is “the degree to which workers … can modify or choose the criteria used for evaluating their performance” (ibid).
Following these theoretical assumptions and empirical evidence, it is suggested that constant connectivity enhances employees’ discretion to redesign work procedures and execute work independently from organisational time and place. Accordingly, it is assumed that constant connectivity leverages employees’ autonomy, particularly through its impact on work method and work scheduling autonomy. This leads to:
Hypothesis 2: Constant connectivity positively affects autonomy.
Autonomy and engagement
Scholars agree that autonomy is an important and positive predictor for wellbeing (Bakker et al., 2011; Hackman, 1976; Hackman & Oldham, 1976; Janz, Colquitt, & Noe, 1997; Kossek et al., 2006; Kossek & Lautsch, 2012; Langfred, 2000, 2005; Ryan & Deci, 2000) and a resource that is functional in stimulating engagement and personal growth
(Bakker & Demerouti, 2007; Bakker et al., 2011). Job resources enhance individuals’ intrinsic motivation and personality integration because they fulfil basic human needs, including the need for autonomy (Bakker et al., 2011). When individuals have greater feelings of self-determination and control (i.e. autonomy), they are ready and inclined towards action that fits the perspective of psychological engagement (Macey & Schneider, 2008).
Greater autonomy motivates individuals to actively shape their role and work context and leads to positive psychological outcomes including increased energy, persistence, vigour and initiative (Spreitzer, 1995). Such resourceful environments also foster the willingness of employees to dedicate their efforts and abilities to the job (Meijman & Mulder, 1998) and to fully invest themselves (Macey & Schneider, 2008). In that regard, autonomy at the
However, engagement also embodies feelings with regard to the involvement of the self (Macey & Schneider, 2008). The more employees can draw on and harness their selves in their work, the more stirring are their levels of absorption (Kahn, 1990). Thus, also the third dimension of engagement is leveraged through increased autonomy, leading to:
Hypothesis 3: Autonomy positively affects engagement.
Indirect effect
Based on the previous line of reasoning, an indirect effect is proposed, i.e. that the relationship between constant connectivity and engagement is indirectly affected by
autonomy. Constant connectivity therein performs the external contextual factor that provides employees greater flexibility as to when, where and how to work and thereby helps them to fulfil one of their innate psychological needs, the need for autonomy (Ryan & Deci, 2000). In that way, employees are motivated to actively shape their role and work context, and increase their willingness to fully dedicate their efforts and abilities to the job.
Additionally, increased autonomy inclines them to harness more of their selves in their work (Macey & Schneider, 2008). All of these outcomes ultimately speak for facilitated
engagement, considering it a persistent affective-motivational state of fulfilment that is characterized by vigour, dedication, and absorption. Lastly, it is thus proposed that:
Hypothesis 4: Constant connectivity positively affects engagement through autonomy.
Methods Participants and sample
The two-wave panel study was conducted at a large Scandinavian natural resources company which was chosen due to its proximity to the researcher. 3,070 knowledge workers were invited to participate in the web-based survey via email at two different times with three months between the measurement points. At time one (T1), 742 employees replied (a 24.2% response rate). Of those, 345 completed the questionnaire also at time two (T2) (a drop-out
rate of 53.5%). Thus, 345 employees completed both waves, resulting in 690 data points and an overall response rate of 11.2%.
The majority of respondents in the final sample was male (65.5%). On average, participants were 42.98 years old (SD = 11.07), with ages from 23 to 75. They worked an average of 10.10 years (SD = 10.91) for the company and reported an average work week of 40.99 hours (SD = 5.96), which was slightly above the contracted amount of 37.5 hours per week that the majority indicated (74.8%). 18.3% of the final sample held a supervisor
position. The scores of those who completed both waves (i.e. the final sample, N = 345) were also compared to the scores of participants that dropped out (N = 397). This revealed that men were slightly overrepresented in the final sample (65.5% were male compared to 60.5% in the initial sample, χ2 = 6.73, p = .009) and that the average age was imperceptibly higher in the final sample (M = 42.98, SD = 11.07) than in the initial sample (M = 42.55, SD = 10.86, t = .99, p = .320). The final sample held a slightly lower average (M = 10.10, SD = 10.91) of organisational tenure than the initial sample at T1 (M = 10.38, SD = 10.83, t = -.42, p = .673), and reported a slightly lower average amount (M = 40.99, SD = 5.96) of actual hours worked per week compared to those of the initial sample (M = 41.31, SD = 7.38, t = -1.12, p = .263). Also, supervisors were slightly underrepresented (18.3% compared to 19.3% in the initial sample, χ2 = .42, p = .515). However, most of these tests were statistically insignificant, so it was assumed that selective drop-outs did not bias the results. Only the participants’ gender resulted in a statistically significant difference, indicating that attention had to be paid to the fact that men were overrepresented in the final sample.
Observed variables
Constant connectivity was measured with five items derived from Büchler et al. (2020). The items focused on staying connected to work during non-worktime by stating for instance, “Through my mobile work device, I know what awaits me at work before I get
there”. Participants indicated their agreement on a 7-point scale, ranging from 1 = strongly disagree to 7 = strongly agree.
Autonomy was measured with nine items derived from Breaugh (1985), as these empirically validated the three distinct facets that scholars broadly discuss (DeCotiis & Koys, 1980; Nicholson, 1984). Participants were asked to indicate their freedom to decide work methods (e.g. “I am allowed to choose the way to go about my job”), work scheduling (e.g. “I have some control over the sequencing of my work activities”) and work criteria (e.g. “I am able to modify what my job objectives are”) on a seven-point scale, ranging from 1 = strongly disagree to 7 = strongly agree.
Engagement was measured with the Utrecht Work Engagement Scale, as it is the most often used scientifically derived measure for the concept (Schaufeli & Bakker, 2010). It included a subscale of three items for each of the engagement dimensions – vigour (e.g. “At my work, I feel bursting with energy”), dedication (e.g. “My job inspires me”) and absorption (e.g. “I am immersed in my work”). Participants indicated their agreement on a seven-point scale, ranging from 0 = never to 7 = always (daily). Table 1 below depicts the descriptive statistics, correlations and reliabilities (alpha coefficients). Table 2 shows all measurement items, the corresponding factor loadings, standard errors, and explained variances.
To control for potential alternative explanations for variances in the dependent variables, the following control variables were included. Gender and age were considered because these correlate with engagement (Bolino, Hsiung, Harvey, & LePine, 2015). Position was included because research depicts that supervisors experience more autonomy than subordinates, which positively impacts their engagement (Gruber et al., 2018). Actual working hours per week were controlled for since they relate to engagement (ten Brummelhuis, Rothbard, & Uhrich, 2017) and correlate with connectivity during non-worktime (Stoner, Stephens, & McGowan, 2009). All control variables were self-reported.
Table 1
Descriptive Statistics, Correlations and Alpha Coefficients
Variables M (SD) 1 2 3 4 5 6 7 8 9 10 Time 1 1. Constant connectivity 4.18 (1.86) .94 2. Autonomy 4.79 (0.96) .18** .88 3. Engagement 5.32 (1.11) .30** .25** .93 Time 2 4. Constant connectivity 4.18 (1.84) .89** .19** .27** .94 5. Autonomy 4.68 (1.08) .25** .69** .27** .29** .90 6. Engagement 5.27 (1.18) .25** .25** .79** .26** .34** .94
Control variables Time 1
7. Gender 1.34 (.48) -.03 .02 .13* -.03 .07 .12* -
8. Age 42.98 (11.07) .12* -.02 .13* .07 -.05 .18** -.19** -
9. Position 1.18 (0.39) .17** .11* .11* .15** .00 .15** -.12* .18** -
10. Working hours per week 40.99 (5.96) .33** .10 .08 .32** .04 .12* -.17** .05 .31** -
Note: N = 345. M = Mean; SD = Standard Deviation. Values on the diagonal in bold are reliabilities (). *p < .05, **p ≤ .001. Position indicates whether the participant was a supervisor or not.
Table 2 Measurement Model Time 1 Time 2 Item R2 St. factor loading Unst. factor loadinga SE R2 St. factor loading Unst. factor loadinga SE Constant connectivity .77 .77
Through my (mobile) work devices, I am always available for my colleagues and/or clients, also during non-work hours.
.60 .774 1.000b .69 .828 1.000b
During non-work hours, I monitor my work through my mobile work device (e.g. checking emails or similar messages).
.87 .953 1.274 .06 .88 .940 1.143 .05
Through my mobile work device, I know what awaits me at work before I get there. .68 .824 1.016 .06 .64 .800 0.921 .05
For me, it is common to check and answer emails or other work-related messages during non-work hours.
.86 .928 1.248 .06 .84 .916 1.122 .05
Through the use of my mobile work device, I stay connected to work during non-work hours.
.84 .919 1.240 .06 .82 .907 1.076 .05
Autonomy .60 .69
Work method autonomy c .70 .838 1.000b .77 .876 1.000b
I am free to choose the method(s) to use in carrying out my work. .79 .891 1.000 .81 .900 1.000
I am allowed to decide how I get my job done. .82 .905 0.970 .04 .83 .908 0.960 .04
I am able to choose the way to go about my job. .79 .887 0.879 .04 .79 .887 .870 .04
Work scheduling autonomy c .62 .790 0.759 .74 .861 0.810
My job is such that I can decide when to do particular work activities. .60 .775 1.000 .63 .794 1.000
I have control over the scheduling of my work. .56 .746 1.028 .08 .63 .791 1.081 .07
I have some control over the sequencing of my work activities. .48 .690 0.728 .06 .68 .825 0.897 .06
Work criteria autonomy c .48 .690 0.733 .55 .739 0.739
I have some control over what I am supposed to accomplish. .65 .806 1.000 .66 .811 1.000
My job allows me to modify the way we are evaluated so that I can emphasize some aspects and play down others.
.42 .647 0.799 .07 .41 .639 0.795 .07
I am able to modify what my job objectives are. .67 .820 1.091 .08 .66 .814 1.091 .08
Engagement .61 .65
At my work, I feel bursting with energy. .65 .804 1.000b .65 .807 1.000b
At my job, I feel strong and vigorous. .72 .847 0.988 .05 .74 .858 1.022 .05
My job inspires me. .55 .739 0.957 .06 .50 .708 0.920 .06
I am proud of the work that I do. .61 .779 0.848 .05 .68 .822 0.940 .05
I feel happy when I am working intensely. .58 .758 0.934 .06 .65 .804 0.944 .05
I am immersed in my work. .48 .691 0.830 .06 .43 .655 0.843 .06
I get carried away when I am working. .35 .593 0.750 .06 .45 .667 0.852 .06
Note: SE = standard error. aAll factor loadings are significant at p < .05. bUnit loading indicator constrained to 1. cValues in this row (bold and italics) represent loadings
on the second-order construct (i.e. autonomy).
Table 3
Fit Statistics for the Study Models
Model Description χ2 df TLI CFI RMSEA 90% CI SRMR Δχ2 (Δdf) Model comparison Mm Measurement model 2042.85 945 0.92 0.93 0.058 0.055; 0.062 0.06
MSR1 Structural model
(M1causal + M2causal)
2068.63 952 0.92 0.92 0.058 0.055; 0.062 0.05
Cross-lagged associations between constant connectivity and autonomy
M1baseline Baseline model 601.57 327 0.96 0.97 0.049 0.043; 0.056 0.06
M1causal Causality model
(M1baseline + CC → Aut.)
596.85 326 0.96 0.97 0.049 0.043; 0.055 0.05 4.72* (1) M1baseline vs. M1causal
M1reversed Reverse causation model
(M1baseline + Aut. → CC)
598.16 326 0.96 0.97 0.049 0.043; 0.055 0.05 3.41 (1) M1baseline vs. M1reversed
M1reciprocal Reciprocal model
(M1causal + M1reversed) 593.56 325 0.96 0.97 0.049 0.043; 0.055 0.05 8.01* (2) 3.29 (1) 4.6* (1) M1baseline vs. M1reciprocal M1causal vs. M1reciprocal M1reversed vs. M1reciprocal
Cross-lagged associations between autonomy and engagement
M2baseline Baseline model 1561.06 567 0.90 0.91 0.071 0.067; 0.076 0.06
M2causal Causality model
(M2baseline + Aut. → Eng.)
1555.77 566 0.90 0.91 0.071 0.067; 0.076 0.06 5.29* (1) M2baseline vs. M2causal
M2reversed Reverse causation model
(M2baseline + Eng. → Aut.)
1558.63 566 0.90 0.91 0.071 0.067; 0.076 0.06 2.43 (1) M2baseline vs. M2reversed
M2reciprocal Reciprocal model
(M2causal + M2reversed) 1553.59 565 0.90 0.91 0.071 0.067; 0.076 0.06 7.47* (2) 2.18 (1) 5.04* (1) M2baseline vs. M2reciprocal M2causal vs. M2reciprocal M2reversed vs. M2reciprocal
Note: df = degrees of freedom; TLI = Tucker-Lewis index; CFI = comparative fit index; RMSEA = root mean square error of approximation; SRMR = standardized root
Data analysis
This study investigates mediation, including two causal relationships (i.e. constant connectivity on autonomy, and autonomy on engagement). To establish causality between two variables, the cause must precede the outcome in time (Holland, 1986). Thus, it focuses on traditional regression-based designs, particularly on linear relationships that enable the examination of changes in individual differences over time (Cole & Maxwell, 2003). The five hypotheses were tested using structural equation modelling in Amos. As only two survey waves were conducted, longitudinal factorial invariance was tested, i.e. “that the relation of the latent variables to the manifest variables is constant over time” (Cole & Maxwell, 2003, p. 570). Here, the assumption is that when statistical properties are constant across waves, Path b in the mediation between M1 and Y2 would be equal to Path b between M2 and Y3 in a
three-wave design. Under this assumption, the Product of ab (Path X1M2 and Path M1Y2) provides
an estimate of the mediational effect of X1 on Y3 through M2 (Cole & Maxwell, 2003). This is
visualized in Figure 1 below.
Figure 1
Path diagrams of cross-sectional and longitudinal models of mediation
Note: This figure is retrieved from Cole and Maxwell (2003, p. 559) and modified for this study. Subscripts
indicate different measurement points. Wave 3 is hypothetically added to visualize the path explanation but was not conducted. X = predictor, M = mediator, Y = outcome.
First, the baseline model (Mbaseline) was investigated by including only autoregressive
associations between the same variables across waves (e.g. constant connectivity at T1 and T2). Afterwards, three more complex models were derived and compared. The causal model (Mcausal) added the hypothesized effects between constant connectivity at T1 and autonomy at
T2, as well as between autonomy at T1 and engagement at T2. The reverse causation model (Mreversed) investigated the opposite relationships from engagement at T1 to autonomy at T2,
and from autonomy at T1 to constant connectivity at T2. Finally, the reciprocal model (Mreciprocal) included all relationships from Mcausal and Mreversed. Different models were
compared through Δχ2 tests and multiple fit indices used to examine the model fit, i.e.
comparative fit index (CFI), Tucker-Lewis index (TLI), standardized root mean square residual (SRMR), and root mean square error of approximation (RMSEA).
Results Measurement model
At both data points, the measurement model including participants’ constant
connectivity, autonomy and engagement depicted acceptable model fit (Hu & Bentler, 2009): χ2 (945) = 2042.85; CFI = 0.93, TLI = 0.92; SRMR = 0.06; and RMSEA = 0.058 (confidence interval CI: 0.055, 0.062). Table 1 shows that all alpha coefficients were above .88,
indicating good reliability of all scales (Bryman, 2016). It also reveals that factor correlations within waves ranged from .18 to .34, indicating discriminant validity, i.e. that the latent constructs were sufficiently distinct. High correlations between the same variables across waves, ranging from .69 to .89, depict that the constructs were relatively stable across time points (i.e. unchanging levels of variables over time, see Table 1). Convergent validity was assessed through the factor loadings and explained variances (see Table 2). All observed items showed significant and sizeable standardized factor loadings between .59 and .95, indicating convergent validity. Each item had an explained variance between .35 and .89 on
their respective first-order latent variable. Accordingly, there were no validity concerns with the measurement model.
Longitudinal factorial invariance
Different levels of longitudinal factorial invariance1 were tested through evaluating
how well increasingly constrained models fitted the data. Weak factorial invariance could be established, Δχ2 (18) = 20.58, p = .301, implying that any changes in the amount of reliable variance among the indicators were adequately captured as changes in the amount of common construct variance (Little et al., 2007). Establishing stricter factorial invariance levels proved more complicated, as factor loadings of six items2 appeared significantly variant across
waves. Nonetheless, the analyses were continued given the following grounds. First, none of these items was problematic in terms of its validity or reliability. Hence, the failure of strong invariance to hold only implied less certainty about the assumptions of the Path b in the mediation, but did not undermine the correctness and consistency of the measures. Second, scholars have stressed that equality constraints consistent with several levels of invariance can perform as most appropriate testing (e.g. Little et al., 2007), which is established here through weak factorial invariance.
Establishing the order of variables
Since this study conducted a two-wave panel study, the hypothesized effects were investigated in three steps: 1) from constant connectivity to autonomy, 2) from autonomy to engagement, and 3) from constant connectivity to engagement through autonomy. First, the cross-lagged associations between constant connectivity and autonomy were examined, while controlling for the effect that variables had on themselves (i.e. the autoregression; Mbaseline).
Table 3 shows all directions of effects that were compared. It depicts that the causal model
1 These included configural, loading, intercept and residual invariance (Little, Preacher, Card, & Selig, 2007). 2 The lack of measurement invariance was mainly caused by the engagement scale, i.e. items 3, 4, 6, and 9. The
(M1causal) including the cross-lagged associations between constant connectivity at T1 and
autonomy at T2 showed the best fit compared to the baseline model (M1baseline); Δχ2 (1) =
4.72, p = .029. The reverse causation model (M1reversed), describing the opposite effect from
autonomy at T1 and constant connectivity at T2 fitted worse than M1baseline; Δχ2 (1) = 3.41, p
= .065 and was thus rejected. Despite showing statistically significant results when compared to M1baseline, the reciprocal model (M1reciprocal) was regarded inferior to M1causal,as it included
the insignificant reversed association and thus was less parsimonious. M1causal was retained.
Second, the cross-lagged associations between autonomy and engagement were examined. Table 3 depicts that again, the causal model (M2causal), including the cross-lagged
correlations between autonomy and engagement, performed better than the baseline model (M2baseline); Δχ2 (1) = 5.21, p = .021, and the reverse causation model (M2reversed) compared to
M2baseline; Δχ2 (1) = 2.43, p = .119. It also fitted better than the reciprocal model (M2reciprocal);
Δχ2 (1) = 2.18, p = .140. Hence, M2causal was chosen as the most parsimonious and
best-fitting model. Lastly, both causal models were combined into the structural model (MSR1) to
test the hypotheses.
Hypotheses Testing
The structural model (MSR1) was run to examine all suggested effects. MSR1 indicated
acceptable model fit (Hu & Bentler, 2009); χ2 (952) = 2068.63; CFI = 0.92; TLI = 0.92; SRMR = 0.05; RMSEA = 0.058 (confidence interval CI: 0.055, 0.062). Figure 2 visualizes all hypothesized effects. Engagement at T1 had no direct effect on constant connectivity at T2 (b = - 0.014. p = .741), neither did the reversed assumption (b = .010, p = .696). Accordingly, both H1a and H1b were rejected. Constant connectivity at T1 however, positively predicted autonomy at T2 (b = 0.072, p = .035), while controlling for reported autonomy at T1. Hence, constant connectivity increased autonomy, so that H2 was retained. Also, autonomy at T1 had a positive effect on engagement at T2 (b = 0.101, p = .022), while controlling for previous
levels of engagement at T1. Thus, H3 was retained. The indirect effect of constant
connectivity on engagement through autonomy was also statistically significant (b = 0.007, CI 95% 0.00, 0.03, p = .048).
All coefficients were very low but explained variances for both dependent variables high. The model explained 75.6% of the variance in autonomy at T2 and 66.8% in the variance of engagement at T2, however these results mainly emerged from autoregressive variances. Autonomy at T1 explained 71.4% of the variance in autonomy in T2, leaving 4.2% to be explained by constant connectivity. Engagement at T1 explained 61.2% of the variance in engagement at T2, leaving 5.6% to be predicted by autonomy. Both variables were stable over time.
Figure 2
Regression model
Wave 1 (T1) Wave 2 (T2)
Note: The figure only visualizes the regression weights between the three latent factors. Observed items and
correlations between the latent factors are not shown for the sake of clarity. Statistical significances are flagged,
Autonomy Autonomy Engagement Engagement 0.101* 0.799** Constant connectivity Constant connectivity 0.989** 0.072* 0.946** 0.010 - 0.014
Discussion
This study investigated the extent to which constant connectivity during non-worktime related to engagement as a persistent affective-motivational state of fulfilment, and considered its potential in enhancing autonomy. One important finding is that connectivity during non-worktime has causal priority over engagement. Hence, the prevailing assumption in literature, i.e. that state engagement precedes desirable work behaviours could not be confirmed.
Second, it demonstrated that constant connectivity facilitates engagement because it enhances autonomy. The role of autonomy is crucial, since the overarching direct relationship between constant connectivity and engagement was not detected. As the analysis revealed, employees’ constant connectivity at T1 had a significant direct effect on their autonomy at T2, and their autonomy at T1 a significant direct effect on engagement at T2. None of the two reversed causation relationships was statistically significant. Accordingly, the causal order of the variables as suggested in the indirect effect is confirmed, implying that engagement is indeed constructed through communicative behaviours, to the extent that these behaviours facilitate autonomy.
Theoretical implications
Although previous research investigated links between constant connectivity and engagement, this study is innovative in that it empirically demonstrates a causal relationship between both concepts and positions autonomy as a positive mediator in the process. The finding that constant connectivity positively affects engagement solely through enhancing autonomy complements and extends theory.
First, the detected causal ordering from constant connectivity to engagement (even though indirect) adds value to other scholars’ findings within the connectivity literature (Cavazotte et al., 2014; MacCormick et al., 2012). While these demonstrated the same order of variables and positive nature of the relationship, their conclusions appeared as corollaries,
since they focused on engagement as a behavioural activity. Connecting during non-worktime can only enhance engagement, if understood as employees’ interactions with technology. In contrast, understanding engagement as an affective-motivational state of fulfilment adds value to the finding because psychologically engaged employees were shown to not only do more (i.e. spend extra time), but to do something different (Kahn, 1990; 1992).
Psychological engagement as an outcome implies that professionals in the investigated corporation connect with their work and address difficult issues (i.e. engagement behaviour), drawing upon all of their skills, abilities and personal resources. In that way, becoming engaged is relevant to these employees personally, which embodies crucial organisational consequences. Psychological engagement typically leads to engagement behaviours that are more desirable for organisations than engagement as a behavioural activity (i.e. spend extra time) because they embody employee development and productivity (Bakker & Demerouti, 2007). While previous studies did not make this important distinction, this study explored engagement as a state and thus provides a solid base to further look into processes that translate this psychological engagement into engagement behaviours.
Second, the detected causal ordering supports the perspective that communication creates individual and organisational realities (Fairhurst & Putnam, 2014) in that constant connectivity indirectly enables employees to shape and construct their engagement. This contradicts much of the traditional engagement literature that assumes engagement as a personality trait or cognitive state that precedes work-related activities and communication (Bakker et al., 2011; Salanova, Schaufeli, Xanthopoulou, & Bakker, 2010). Yet, it conforms to van Zoonen and Banghart (2018) who “reversed the traditional figure-ground relationship between engagement and everyday work practices”, showing that work-related social media communication precedes and helps the construction of employee engagement (p. 288). Van Zoonen and Banghart (2018) however, solely investigated direct effects between the
variables, leaving underlying mechanisms unexplored. Here, this was addressed through adding autonomy as a positive mediator in the process – a crucial consideration, since it turned out that constant connectivity only facilitated engagement through leveraging employees’ autonomy.
Third, and even more importantly, the detected causal relationship advances our thinking about what employees’ connectivity behaviours during non-worktime are and how they should be viewed in organisations. The findings suggest that constant connectivity is not only an inherent feature of today’s workplaces but can be viewed as a functional resource (Piszczek, 2017; Sonnentag, 2017). This is important because resources are aspects of a job that help employees achieve work goals, reduce psychological and physiological costs and stimulate growth and development (Schaufeli & Bakker, 2004). As such, they are not only necessary to deal with job demands or to ‘get things done’, but they are also important in their own right or as conduits to the achievement or protection of other valued resources (Hobfoll, 2002). Conceptualizing constant connectivity as a functional resource is a perspective that shifts the figure-ground relationships between constant connectivity and work outcomes that typically view connectivity as a demand of contemporary workplaces leading to detrimental outcomes such as work-life conflict and exhaustion (e.g. Boswell & Olson-Buchanan, 2007; Büchler et al., 2020).
It should be noted that constant connectivity does not directly enhance engagement. However, this is in line with other scholars who suggest that constant connectivity might not directly link to wellbeing and performance but rather interact with other job resources or demands (Sonnentag, 2017). In this study, connectivity was not only needed for professionals to succeed in their work for a corporation with global teams, but portrayed a means to fulfil and leverage their need of autonomy, which ultimately, facilitated a persistent
affective-motivational state. Hence, connectivity operated as a conduit to achieve or protect other valued resources (i.e. autonomy) and ultimately engagement.
This process can also be explained by the conservation of resources theory (Hobfoll, 2001) that emphasizes that resources do not exist individually but travel in caravans and are impacted by conditions that either foster, or limit resource creation and sustenance (Hobfoll, Halbesleben, Neveu, & Westman, 2018). Here, the positive process dominated – constant connectivity nurtured other resources, i.e. autonomy and indirectly, constructed engagement. Yet, this finding needs to be further embedded, considering the figure-ground perspective that classifies implications of constant connectivity more negatively.
Again, conservation of resources theory is helpful because it posits two cycles; resource gains and resource losses, both with a spiralling nature. Resource losses are
disproportionately more salient and rapid than resource gains because they create stress and reduce resources that help in offsetting further losses (Hobfoll, 2001). Resource gains in contrast, are weaker and develop slower, which makes them more difficult to detect (Hobfoll, 2001). Yet, paradoxically, gains become more salient when loss circumstances are high, which means that the motivation to build resource gain cycles increases when losses occur (Hobfoll et al., 2018).
This suggests that in the present study a resource gain cycle stood out, which typically develops slower and weaker than a resource loss and might thus not be prevalent in cross-sectional studies. However, it might have also emerged because of a context, in which (threat of) resource loss is also high, something this study did not consider, while the figure-ground perspective might have focused explicitly on resource losses. This suggests to direct more attention to both processes that can emerge through constant connectivity, which will be further detailed below. Nonetheless, it was demonstrated here that there are circumstances in which constant connectivity facilitates engagement, to the extent that it operates as a resource
and conduit for autonomy. This makes it worthwhile to reconsider what connectivity behaviours are.
Practical implications
The finding that constant connectivity can play a motivational role through which employees experience more autonomy and, ultimately, benefit from higher psychological engagement, is also relevant for organisations. Many organisations focus on limiting
connectivity during non-worktime to prevent negative consequences for their employees, such as burnout, work-life conflict, and escalating engagement. Extreme cases are, for instance, Volkswagen and Daimler stopping its servers from sending emails during non-worktime (Haridy, 2018), and BMW and Deutsche Telekom following with similar policies (Strangler, 2015). Although such policies might mitigate negative consequences of communication technology use, the findings here suggest that disconnecting employees would equally deprive them from the potential good that constant connectivity does.
Rather, organisations need to support employees in reaching and maintaining an optimal state of connectivity with less radical measures, such as communicating that there is no expectation that emails are answered after a certain time. Managing expectations regarding employees’ responsiveness after-hours would enable employees to ‘switch off’ with ease, particularly those that face difficulties to mentally detach from work or feel pressured to react immediately to incoming emails (Barley et al., 2011) or calls (Bordi et al., 2018). Thereby, such policies would actively support employees’ recovery time that was proven important for enhanced wellbeing (Sonnentag, 2003; Sonnentag & Bayer, 2005), and also be viable as they are under managerial control. Several governments have already introduced laws to make such managing of expectations legally binding for employers (Haridy, 2018).
Shifting towards a results-oriented work environment where the quality of work achieved matters more than the presence of employees – whether physically or through
MWDs – would be a crucial step (Kossek, 2016). This is particularly relevant against the background of the current Covid-19 pandemic, in which whole economies have to re-invent themselves and remote work (might) become the ‘new normal’. This will unavoidably increase the dependency on communication technologies (Ofcom, 2020). Finding ways to utilize connectivity to boost employee engagement will thus be vital for organisations seeking their ways out of the crisis. While this study may not provide concrete insights in how to do that, it does emphasize the importance of autonomy as a perpetual job resource, as well as the notion that communicative behaviour is proven to facilitate psychological engagement. It shows that connectivity is not necessarily something to worry about but instead an
opportunity. Those organisations that manage to create processes through which employees can harness connectivity for good, will continue to enjoy competitive advantage because their employees do not only do ‘more’, but do things ‘smarter’ (Macey & Schneider, 2008).
Limitations and future research
In terms of the set-up of the study, there are a few avenues to improve. First, the study relies on two survey rounds, while three waves would be beneficial when testing mediation. Usually, an independent pair of waves is used to assess the effects between the predictor and the mediator, as well as the mediator and the outcome (Cole & Maxwell, 2003). However, in two-wave studies establishing longitudinal factorial invariance serves its purpose (Cole & Maxwell, 2003) and was accordingly done here. Also, data was gathered with only a three months’ time gap. While this helped to safeguard employees’ participation, it might have resulted in small effect sizes from the predictors. Engagement as a psychological ‘mindset’ is a relatively enduring state that serves to explain persistence among other indicators (Macey & Schneider, 2008). Thus, its individual measures are relatively stable over time which was seen in this study, given large autoregressions – also for autonomy, another perpetual concept. Future research would thus benefit from more measurement points over longer time intervals.
In terms of substance, this study adds a perspective on constant connectivity being a conduit for employees to leverage their autonomy, and ultimately facilitate their engagement. Yet, the interaction between constant connectivity and autonomy requires further elaboration. Here, autonomy was measured as an employee’s discretion to choose on work methods, scheduling and criteria, while not accounting for its counterpart, i.e. obligation. Thus, it remains unclear whether employees simply rationalize their connectivity behaviours as their own choice, while actually limiting their autonomy through their behaviours as Mazmanian et al. (2013) suggest. Since studies show that external contexts can equally translate into
controlled motivational factors that pressure individuals to do something (Mazmanian et al., 2013; Ryan & Deci, 2000) and diminish their wellbeing (Büchler et al., 2020), future research would benefit from measuring both, autonomy and obligation.
It could add individual antecedents such as personality traits and boundary preferences (Gruber et al., 2018) or other organisational factors including norms and expectations
(Richardson & Benbunan-Fich, 2011), or communication control (ter Hoeven & van Zoonen, 2020). All of these were shown to also reduce employees’ control over their work conditions (Mazmanian & Erickson, 2014) and to negatively impact coping strategies with connectivity (Gruber et al., 2018). In that way, future research could not only test the presence of the so-called ‘autonomy paradox’ (Mazmanian et al., 2013) but also spell out how and when the positive experience of connectivity (i.e. as a resource) turns into a demanding one.
This directly relates to another area for future research – the question of how much connectivity is desirable. Though often referred to as constant, connectivity is seldom, if ever, constant. Individuals are still agentic in switching between connects and disconnects (Russo et al., 2019) and can adjust their connectivity levels to the situation at hand (Kolb et al., 2012). As connectivity is here to stay, future research needs to characterize states and patterns of connectivity because this will help to further understand the impact of too much or too
little connectivity, and ultimately the ‘right’ amount for favourable coping strategies as investigated here. Albeit these limitations, this study has shown that it is worthwhile and timely to rethink constant connectivity behaviours of employees – not only for academia but also for business as the need for flexible and remote work practices will continue. The recent Covid-19 pandemic and remote work mandates have made this crystal-clear and further accelerated this ongoing trend, relying on a certain level of connectivity.
References
Bakker A. B., Albrecht, S. L., & Leiter, M. P. (2011). Key questions regarding work
engagement. European Journal of Work and Organizational Psychology, 20(1), 4-28. doi: 10.1080/1359432X.2010.485352
Bakker, A. B., & Demerouti, E. (2007). The Job Demands-Resources Model: State of the Art. Journal of Managerial Psychology, 22(3), 309–328. doi: 10.1108/02683940710733115 Barley, S. R., Meyerson, D. E., & Grodal, S. (2011). E-mail as a source and symbol of stress.
Organization Science, 22(4), 887-906. doi: 10.1287/orsc.1100.0573
Bolino, M. C., Hsiung, H. H., Harvey, J., & LePine, J. A. (2015). Well, I’m tired of Tryin’! Organizational Citizenship Behavior and Citizenship Fatigue. Journal of Applied Psychology, 100(1), 56-74. doi:10.1037/a0037583
Bordi, L., Okkonen, J., Mäkiniemi, J. P., & Heikkilä-Tammi, K. (2018). Communication in the Digital Work Environment: Implications for Wellbeing at Work. Nordic Journal of Working Life Studies, 8(53), 29-48. doi: 10.18291/njwls.v8iS3.105275
Boswell, W. R., & Olson-Buchanan, J. B. (2007). The use of communications technologies after hours: The role of work attitudes and work-life conflict. Journal of Management, 33(4), 592–610. doi:10.1177/0149206307302552
Breaugh, J. A. (1985). The measurement of Work Autonomy. Human Relations, 38(6), 551-570. doi: 10.1177/001872678503800604
Brown, S. P. (1996). A Meta-Analysis and Review of Organizational Research on Job Involvement. Psychological Bulletin, 120(2), 235–255. doi: 10.1037/0033-2909.120.2.235
Bryman, A. (2016). Social Research Methods. Oxford, United Kingdom. Oxford University Press.
Büchler, N., ter Hoeven, C. L., & van Zoonen, W. (2020). Understanding Constant Connectivity to Work: How and for Whom is Constant Connectivity Related to Employee Well-Being. Information and Organization, 30(3). doi:
10.1016/j.infoandorg.2020.100302
Cavazotte, F., Lemos, A. H., & Villadsen, K. (2014). Corporate smart phones: professionals’ conscious engagement in escalating work connectivity. In: New Technology, Work and Employment, 29(1), 72-87. doi: 10.1111/ntwe.12022
Chesley, N. (2005). Blurring Boundaries? Linking Technology Use, Spillover, Individual Distress, and Family Satisfaction. Journal of Marriage and Family, 67(5), 1237-1248. doi: 10.1111/j.1741-3737.2005.00213.x
Chesley, N. (2006). Families in a high-tech age: Technology usage patterns, work and family correlates, and gender. Journal of Family Issues, 27(5), 587–608. doi:
10.1177/0192513X05285187
Cole, D. A., & Maxwell, S. E. (2003). Testing Models With Longitudinal Data: Questions and Tips in the Use of Structural Equation Modeling. Journal of Abnormal Psychology, 112(4), 558-577. doi: 10.1037/0021-843X.112.4.558
DeCotiis, T. A., & Koys, D. J. (1980). The identification and measurement of the dimensions of organizational climate. Academy of Management Proceedings, 31, 171-175. doi: 10.5465/ambpp.1980.4976195
Derks, D., & Bakker, A. B. (2010). The impact of e-mail communication on organizational life. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 4(1).
Dery, K., & MacCormick, J. (2012). Managing Mobile Technology: The Shift from Mobility to Connectivity. MIS Quarterly Executive, 11(4), 159-173.
Fairhurst, G. T., & Putnam, L. L. (2004). Organizations as discursive constructions. Communication Theory, 14(1), 5–26. doi:10.1111/j.1468-2885.2004.tb00301.x Fairhurst, G. T., & Putnam, L. L. (2014). Organizational discourse analysis. In L. L. Putnam
& D. K. Mumby (Eds.), The SAGE handbook of organizational communication:
Advances in theory, research, and methods (3rd ed., pp. 271–296). Thousand Oaks, CA: SAGE Publications, Inc.
Gagne, M., & Deci, E. L. (2005). Self-determination theory and work motivation. Journal of Organizational Behavior, 26, 331–362.
Gruber, M. R., Sarigianni, C., Geiger, M., & Remus, U. (2018, January). Do You Plead Connectivity? – Understanding How Lawyers Deal With Constant Connectivity. Paper
presented at the 51st Hawaii International Conference on System Sciences, Hawaii, USA. Retrieved from http://hdl.handle.net/10125/50545.
Holland, P. W. (1986). Statistics and causal inference. Journal of the American Statistical Association, 81(396), 945–960. doi: 10.1080/01621459.1986.10478354
Hackman, J. R. (1976). Group influences on individuals. In: M. Dunnette (Eds.), Handbook of Industrial Organizational Psychology (pp.1455-1526). Rand McNally, Chicago.
Hackman, J. R. (1980). Work redesign and motivation. Professional Psychology, 11(3), 445-455. doi: 10.1037/0735-7028.11.3.445
Hackman, J. R., & Oldham, G. R. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60(2), 159-170. doi: 10.1037/h0076546
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250-279. doi: 10.1016/0030-5073(76)90016-7
Haridy, R. (2018, August 14). The right to disconnect: The new laws banning after-hours work emails. New Atlas. Retrieved from https://newatlas.com/right-to-disconnect-after-hours-work-emails/55879/
Hobfoll, S. E. (2001). The influence of culture, community, and the nested-self in the stress process: advancing conservation of resources theory. Applied Psychology: An
International Review, 50(3), 337-70. doi: 10.1111/1464-0597.00062
Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of General Psychology, 6(4), 307-324. doi: 10.1037/1089-2680.6.4.307
Hobfoll, S. E., Halbesleben, J., Neveu, J.-P., & Westman, M. (2018). Conservation of Resources in the Organizational Context: The Reality of Resources and Their Consequences. Annual Review of Organizational Psychology and Organizational Behavior, 5, 103-128. doi: 10.1146/annurev-orgpsych-032117-104640
Hu, L., & Bentler, P. M. (2009). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1-55. doi: 10.1080/10705519909540118
Janz, B. D., Colquitt, J. A., & Noe, R. A. (1997). Knowledge worker team effectiveness: The role of autonomy, interdependence, team development and contextual support variables. Personnel Psychology, 50(4), 877-904. doi: 10.1111/j.1744-6570.1997.tb01486.x Kahn, W. A. (1990). Psychological conditions of personal engagement and disengagement at
Kahn, W. A. (1992). To be fully there: Psychological presence at work. Human Relations, 45(4), 321–349. doi: 10.1177/001872679204500402
Kelliher, C., & Anderson, D. (2010). Doing more with less? Flexible working practices and the intensification of work. Human Relations, 63, 83-106. doi:
10.1177/0018726709349199
Kolb, D. G., Caza, A., & Collins, P. D. (2012). States of connectivity: New Questions and New Directions. Organization Studies, 33(2), 267-273. doi:
10.1177/0170840611431653
Kossek, E. E. (2016). Managing work-life boundaries in the digital age. Organizational Dynamics, 45(3), 258-270. doi: 10.1016/j.orgdyn.2016.07.010
Kossek, E. E., & Lautsch, B. A. (2012). Work–family boundary management styles in
organizations: A cross-level model. Organizational Psychology Review, 2(2), 152–171. doi: 10. 1177/2041386611436264
Kossek, E. E., Lautsch, B. A., & Eaton, S. C. (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. doi:
10.1016/j.jvb.2005.07.002
Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. (2009). Balancing borders and bridges: Negotiating the work-home interface via boundary work tactics. Academy of Management Journal, 52(4), 704–730. doi:10.5465/AMJ.2009.43669916
Langfred, C. W. (2000). The paradox of self-management: Individual and group autonomy in work groups. Journal of Organizational Behaviour, 21(5), 563-585. doi: 10.1002/1099-1379
Langfred, C. W. (2005). Autonomy and performance in teams: The multilevel moderating effect of task interdependence. Journal of Management, 31(4), 513-529. doi:
10.1177/0149206304272190
Little, T.D., Preacher, K. J., Card, N. A., & Selig, J. P. (2007). New developments in latent variable panel analyses of longitudinal data. International Journal of Behavioral Development, 31(4), 357-365. doi: 10.1177/0165025407077757
MacCormick, J. S., Dery, K., & Kolb, D. G. (2012). Engaged or just connected? Smartphones and employee engagement. Organizational Dynamics, 41(3), 194-201. doi:
10.1016/j.orgdyn.2012.03.007
Macey, W. H., & Schneider, B. (2008). The meaning of employee engagement. Industrial and Organizational Psychology, 1(1), 3–30. doi:10.1111/j.1754-9434.2007.0002.x
Mazmanian, M. (2013). Avoiding the trap of constant connectivity: When congruent frames allow for heterogeneous practices. Academy of Management Journal, 56(5), 1225-1250. doi: 10.5465/amj.2010.0787
Mazmanian, M., & Erickson, I. (2014, April). The Product of Availability: Understanding the Economic Underpinnings of Constant Connectivity. Paper presented at the CHI 2014, Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Toronto, Canada. Retrieved from https://dl.acm.org/doi/10.1145/2556288.2557381 Mazmanian, M., Orlikowski, W. J., & Yates, J. (2013). The autonomy paradox: The
implications of mobile email devices for knowledge professionals. Organization Science, 24(5), 1337-1357. doi: 10.1287/orsc.1120.0806
Meijman, T. F., & Mulder, G. (1998). Psychological aspects of workload. In P. J. Drenth, H. Thierry, & C. J. de Wolff (Eds.), Handbook of Work and Organizational Psychology (pp. 5-33). Erlbaum, Hove.
Nicholson, N. (1984). A theory of work role transitions. Administrative Science Quarterly, 29(2), 172-191. doi: 10.2307/2393172
Ofcom (2020, June 24). Online Nation 2020 Report. Ofcom Publications. Retrieved from https://www.ofcom.org.uk/__data/assets/pdf_file/0027/196407/online-nation-2020-report.pdf
Perlow, L. A. (2012). Sleeping with Your Smartphone: How to Break the 24/7 Habit and Change the Way You Work. Boston, MA: Harvard Business Review Press.
Piszczek, M. M. (2017). Boundary control and controlled boundaries: Organizational
expectations for technology use at the work-family interface. Journal of Organizational Behavior, 38, 592-611. doi:10.1002/job.2153
Richardson, K., & Benbunan-Fich, R. (2011). Examining the antecedents of work
connectivity behavior during non-worktime. Information and Organization, 21(3), 142-160. doi: 10.1016/j.infoandorg.2011.06.002
Rothbard, N. P., & Edwards, J. R. (2003). Investment in work and family roles: A test of identity and utilitarian motives. Personnel Psychology, 56(3), 699–730. doi: 10.1111/j.1744-6570.2003.tb00755.x
Russo, M., Ollier-Malaterre, A., & Morandin, G. (2019). Breaking out from constant
connectivity: Agentic regulation of smartphone use. Computers in Human Behavior 98, 11-19. doi: 10.1016/j.chb.2019.03.038
