Coping with pandemics using social network sites: A psychological detachment perspective to COVID-19 stressors

Abstract

Prior research has often portrayed information technology (IT) as a stressor. In this paper, we propose and demonstrate that IT can also be an effective means of coping with life stressors, including those induced by pandemics such as COVID-19. We thus deviate from the common IT-as-a-stressor perspective and adopt an IT-as-a-coping-mechanism viewpoint. To this end, we apply the stressor-detachment model from organisational psychology to the use of social network sites (SNSs) in coping with stressors wrought by the COVID-19 pandemic. We examine psychological well-being as our dependant variable and introduce psychological detachment through SNS use as a mediator and moderator of the associations between psychological well-being and two COVID-19 stressors: work–family conflict and perceived isolation. We used structural equation modelling and tested this model with survey data collected from 398 professionals who were in lockdown and working from home during the pandemic. The results indicated that psychological detachment through SNS uses increased psychological well-being and that heightened work–family conflict motivated this detachment strategy. In contrast, consistent with helplessness and motivation–opportunity theories, perceived isolation as a stressor did not influence psychological detachment through SNS use. While perceived isolation directly reduced individual well-being, the effect of work–family conflict on well-being was contingent upon users' levels of psychological detachment through SNS use. These findings suggest that while psychological detachment through SNS use is an effective means of improving one's well-being, it can be positively or negatively affected by stressors. Our study contributes to research on technology-mediated strategies for coping with stress and the psychosocial implications of global pandemics.

1. Introduction

The COVID-19 pandemic dramatically decreased people's face-to-face social interactions and increased the practice of working remotely (Brynjolfsson et al., 2020). The decline in social interactions can drive social isolation (cf. Marshall et al., 2007), whereas the upswing in remote work can drive work–family conflict by blurring the boundaries between work and family time (cf. Boswell and Olsen-Buchanan, 2007). Research has empirically documented the prevalence of social isolation and work–family conflict as stressors during the pandemic (Carnevale and Hatak, 2020; Bourizi et al., 2020; Songsangyos and Iamamporn, 2020).

Social isolation and work–family conflict as stressors are not, however, pandemic-specific; the literature has extensively documented their detrimental effects on health and well-being (Anderson et al., 2002; Steptoe et al., 2013). For example, scholars have linked social isolation to depressive symptoms amongst older adults (Cacioppo et al., 2010). Nevertheless, the pandemic, lockdown and associated changes in work routines, including the blurring of boundaries between work and free time, have the potential to engender and exacerbate such stressors (Carnevale and Hatak, 2020; Bourizi et al., 2020; Songsangyos and Iamamporn, 2020). Social isolation is a universal stressor in both humans (Grant et al., 2009) and animals (Weiss et al., 2004) because it represents an infringement on the basic need to socialise (Chao et al., 1994). Similarly, work–family conflict is an important stressor (Noor and Maad, 2008) because it creates adverse and demanding situations for employees who are working from home—i.e. a context in which family demands interfere with work duties and work duties interfere with family demands (Shimazu et al., 2013). Importantly, such stressors can have a detrimental effect on psychological well-being (Noor and Maad, 2008). During the pandemic, people utilised information technology (IT), such as social networking sites (SNSs) and video games, to cope with such stressors because SNSs help to pass the time and enable individuals to interact with others (i.e. socialise) and mentally disengage or detach from the stress caused by COVID-19 (Cinelli et al., 2020). Thus, it is hardly surprising that the use of Facebook hit record levels during the COVID-19 pandemic.1 In this paper, we scrutinise psychological detachment, i.e. temporarily mentally disengaging and refraining from activities or events that cause stress, as a key mechanism that SNS use affords to cope with COVID-19 stressors. Our argument is grounded in prior research on work and organisational psychology, which suggests detachment as an effective means of preventing the adverse effects of stressors on well-being (Sonnentag et al., 2010; Sonnentag and Kruel, 2006). Thus, we argue that in the context of a global pandemic, the desire to detach oneself from the stress related to social isolation and challenges in the work–family balance may motivate SNS use.

The current study aims to address a major lacuna in the research on IT-mediated strategies for coping with stressors (e.g. D'Arcy et al., 2014; Tarafdar et al., 2020). Most prior literature has examined coping behaviours in response to the demands of a technology that an individual considers troubling (e.g. Bala and Venkatesh, 2016; Stein et al., 2015; Tarafdar et al., 2020). In other words, prior research has largely portrayed technology—and the changes and glitches therein—as a stressor (Ragu-Nathan et al., 2008; Tarafdar, Pirkkalainen et al., 2019; Salo et al., 2020). Moreover, prior research has shown that SNS use can itself be a source of stress (C. Maier, Laumer, Eckhart and Weitzel, 2012; C. Maier et al., 2013; C. Maier et al., 2015; C. Maier et al., 2015). However, individuals can also use IT to cope with non-IT stressors, as we later explain. Thus, we depart from the dominant line of reasoning without discounting its importance. We specifically argue that IT use can enable psychological detachment from stressors and thus possibly help individuals to cope with stressful events, such as the COVID-19 pandemic, which are not essentially IT-related. In doing so, we adopt an IT-as-solution perspective rather than the dominant IT-as-stressor perspective and address the following research question:

To what extent can SNS use promote psychological detachment and help individuals to cope with COVID-19 stressors?

To answer this question, we applied the stressor-detachment model (Sonnentag and Fritz, 2015) to the COVID-19 pandemic and SNS use. This contextualisation includes work–family conflict (Boswell and Olsen-Buchanan, 2007) and perceived isolation (Marshall et al., 2007) as key stressors created by COVID-19, psychological detachment through SNS use as the recovery process and psychological well-being as the outcome. Our focus on isolation and work–family conflict as stressors is guided by findings that both isolation (Clay and Parker, 2020; Torales et al., 2020; Van Bavel et al., 2020) and work–family conflict (Fessell and Cherniss, 2020; Restubog et al., 2020; Songsangyos and Iamamporn, 2020) have functioned as key sources of stress during the pandemic. We focus on psychological well-being as the outcome because detachment research frequently utilises it as a manifestation of reduced strain (Sonnentag et al., 2010; Sonnentag and Kruel, 2006). Following the stressors–detachment model (Sonnentag and Fritz, 2015), we suggest that psychological detachment through SNS use has both mediating and moderating roles in the relationships between stressors and strain.

We tested our research model with data from 398 professionals who were in confinement due to the COVID-19 pandemic. We analysed the data using structural equation modelling. The results showed that psychological detachment through SNS use was positively associated with psychological well-being and that perceived isolation had a negative direct effect on psychological well-being.The results also indicated that psychological detachment through SNS use amplified the negative effect of work–family conflict on psychological well-being. Together, the findings indicate the ineffectiveness of psychological detachment through SNS use as a coping strategy for alleviating the effects of our two stressors associated with the COVID-19 pandemic—namely, isolation and work–family conflict.

By investigating psychological detachment through SNS use as a potential coping strategy for COVID-19 stressors, our paper contributes to the literature on psychological detachment (Sonnentag et al., 2010; Sonnentag and Kruel, 2006), which has not yet focused on IT as a vehicle for detachment, as well as the literature on technology-mediated coping mechanisms for managing stressors (e.g. D'Arcy et al., 2014; Tarafdar et al., 2020) and the emerging body of literature on the psychosocial implications of pandemics (e.g. Ågerfalk et al., 2020; Laato et al., 2020; Rai, 2020).

The remainder of the paper is organised as follows. After this introductory section, we describe the theoretical background in Section 2. The third section is dedicated to hypothesis development. In Section 4, we describe our research methodology and data analysis results. In the fifth section, we discuss the key findings as well as their theoretical and practical implications. A discussion of the limitations and avenues for future research follows in the sixth section. Finally, Section 7 concludes the paper.

2. Theoretical background

2.1. Technology-mediated coping with stressors

The transaction-based approach (Lazarus, 1966; McGrath, 1976; Lazarus and Folkman, 1984; Cooper et al., 2001) has provided the foundation for several studies on stress. In brief, the transaction-based approach describes stress as the combination of a stimulating condition and an individual's response to it. Thus, according to the person-environment (P–E) theory, stress is a psychological reaction to an imbalance between the person and the environment (e.g. Cooper et al., 2001). Stressors are events, demands, stimuli or conditions in the work/organisational environment that cause individuals to experience stress (Cartwright and Cooper, 1997). Strain, in turn, refers to the observable behavioural, psychological and physiological outcomes of stress at an individual level (Kahn and Byosiere, 1992; Cooper et al., 2001).

According to Lazarus (1966) and Lazarus and Folkman (1984), people respond to stressors by engaging in coping behaviours. Coping is conceptualised as ‘cognitive and behavioural efforts to manage specific external and/or internal demands that are appraised as taxing or exceeding the resources of the person’ (Folkman and Lazarus, 1984, p. 141). Individuals triggered by stressors can utilise two general coping strategies—problem-focused coping and emotion-focused coping (Lazarus, 1966; Lazarus and Folkman, 1984). Problem‐focused coping entails efforts to alter or manage stressful situations. In contrast, emotion‐focused coping involves changing the way one thinks or feels about a stressful situation. Emotion-focused coping also involves cognitive processes, such as reappraisals and disengagement.

The prior literature on coping has focused predominantly on coping behaviours in response to demands associated with the implementation of new IT in the workplace or glitches in existing IT that the individual considers troubling (see Tarafdar et al., 2020). These coping behaviours are responses to a person's perceptions of discrepant features in the new system—for example, regarding task uncertainty resulting from the changes in work processes following the implementation of a new IT. People respond to stressful events and situations using several coping actions, such as venting, seeking social support and distancing themselves psychologically from the troubling situation (Beaudry and Pinsonneault, 2010; Stein et al., 2015). Their actions can also include IS use-related behaviours, such as resistance or avoidance, efforts to limit usage to the minimum level required, experimenting with fewer features to execute existing tasks and work processes (Bala & Venkatesh, 2015; Beaudry and Pinsonneault, 2005; de Guinea and Webster, 2013) and altering tasks to increase their compatibility with the new system (Beaudry and Pinsonneault, 2005). Table 1 provides an overview of key studies on coping behaviours in the prior literature.

Table 1.

Examples of studies on coping with IT-induced stress.

Study	Coping strategy	Data and research context	Theory	Key findings
Tarafdar et al. (2020)	Distraction (within and outside SNS)	A three-wave survey of 444 Facebook users	Concept of feature-rich IT, theory of technology frames and distraction as a coping behaviour	SNS stressors influence distraction within and outside SNS. In turn, distraction within SNS predicts addiction.
Pirkkalainen et al. (2019)	IT control, positive reinterpretation, distress venting, distancing from IT	A cross-sectional survey study of 846 organisational IT users	Technostress and proactive and reactive coping	The reactive coping behaviours of distress venting and distancing from IT can alleviate technostress by decreasing the negative effect of technostress creators on IT-enabled productivity. The proactive coping behaviours of positive reinterpretation and IT control can help IT users by increasing the effectiveness of reactive coping behaviours and contributing to IT-enabled productivity.
Tafardar et al. (2019)	Venting, distancing, positive IT outlook, IT use skills, IT use autonomy; time-related demarcations, work and non-work IT use separation	A qualitative study in the UK involving interviews of 30 executives/knowledge workers and a survey of 846 US employees who use IT in their workplace		Guidelines for organisational policy and action include informing and educating employees, identifying the fit between employees and various technostress coping behaviours and encouraging employees to develop personal technostress coping strategies.
C. Maier et al. (2015b)	Discontinuous SNS usage	An experiment involving 82 student Facebook users	IS continuance and technostress literature	SNS-stress creators and SNS exhaustion cause discontinuous usage intentions. Switching- stress creators and switching exhaustion reduce discontinuous usage intentions.
Galluch et al. (2015)	Method control and resource control	Two laboratory experiments (n = 90 in each experiment) with university students	Transactional model of stress	ICT-enabled demands serve as stressors and lead to perceptual stress. ICT-enabled timing control negatively moderates the relationship between stressors and stress. Method control negatively moderates the relationship between perceptual conflict and strain while increasing the relationship between perceptual overload and strain. Resource control has the opposite effects.
D'Arcy et al. (2014)	Moral disengagement	Survey of 539 employees	Coping theory and moral disengagement theory	Security-related stress impacts moral disengagement, which, in turn, impacts information security policy violation intention.
Bala & Venkatesh (2015)	Technology adaptation behaviours (avoidance, exploitation, exploration‐to‐innovate and exploration‐to‐revert)	Two field studies (N = 211 and N = 181) with four-wave data collection in two organisations	Transactional model of stress and coping and coping model of user adaptation	Technology adaptation behaviours influence changes in job performance and job satisfaction.
de Guinea & Webster (2013)	Exploitive and adaptive behaviours	Experience sampling study and experiment with 58 employees.	Conceptualisations of IS use, coping theory affect–object paradigm and automaticity perspective	Users engage in two main IS use patterns—automatic and adjusting. The automatic IS use pattern takes place during expected IT events, whereas the adjusting IS use pattern is triggered by discrepant IT events.
Stein et al. (2015)	Venting, adapting and seeking instrumental support	Semi-structured interviews with 47 university employees	Conceptualisations of IS use, emotions and IT use	Users respond to uniform emotions with clear adaptation strategies. Users respond to ambivalent emotions by combining various adaptation strategies.
Beaudry & Pinsonneault (2010)	Venting, distancing, seeking social support, engaging in task adaptation and seeking instrumental support	Survey of 249 bank account managers	Coping model of user adaptation and coping theory	Anger is related to IT use via the seeking of social support. Anxiety is related to IT use directly and via distancing and the seeking of social support. Happiness and excitement are related to IT use via task adaptation.
Beaudry & Pinsonneault (2005)	Four adaptation strategies (benefits satisficing, disturbance handling, self‐preservation and benefits maximisation)	Interviews with six account managers	Coping theory	The paper proposes the coping model of user adaptation by suggesting that the four adaptation strategies result in three outcomes: restoring emotional stability, minimising the perceived threat of technology and improving user effectiveness and efficiency.
Lee & Larsen (2009)	Response efficacy, self-efficacy and perceived costs	Survey of 239 US executives	Protection motivation theory	Threats and coping appraisals predict anti-malware software adoption intentions.

The COVID-19 pandemic created various stressors. Policies restricting travel outside the home and the use of public services dramatically increased the number of people working from home. In addition, school closures arguably increased the demands of family life for parents. The lockdown, which was implemented to prevent the pandemic's spread, also introduced social isolation and thus increased loneliness.2 In this paper, therefore, we focus on the stress caused by work–family conflict (Bonebright et al., 2000; Boswell and Olsen-Buchanan, 2007) that resulted from working from home and the perceived isolation (Marshall et al., 2007) that resulted from measures to fight the COVID-19 pandemic.

2.2. Stressor-detachment model

In the current study, we explore whether SNS use can serve as a means of psychologically detaching oneself from stressors wrought by the COVID-19 pandemic. To this end, we employ the stressor-detachment model (Sonnentag and Fritz, 2015; see Fig. 1 ) from work and organisational psychology as the theoretical backbone of the study. Drawing on the cognitive activation theory of stress (Ursin and Eriksen, 2004) and the allostatic load model (Ganster and Rosen, 2013), Sonnentag and Fritz (2015, p. S75) asserted that “‘it is not primarily the acute stress reaction that is detrimental for an organism but rather the sustained activation, even when the stressor is no longer present”. In brief, the stressor-detachment model postulates that psychological detachment from work during non-work time is an important emotion-focused strategy for coping with work-related stressors.

Fig. 1.

Stressor-detachment model (Sonnentag and Fritz, 2015).

The stressor-detachment model is appropriate for this study for two reasons. First, the prior literature has primarily employed the coping theory and its variants to investigate stress and coping strategies (e.g. Beaudry and Pinsonneault, 2005, 2010; D'Arcy et al., 2014). Although the stressor-detachment model has been widely used in the organisational behaviour literature (e.g. Kinnunen and Feldt, 2013; Sonnentag and Fritz, 2007, 2015), very few, if any studies, have applied the model to examine the effects of digital technology. By applying the stressor-detachment model, our research extends this body of work and increases the diversity of theoretical perspectives applied to the study of digital technology. Second, the stressor-detachment model provides a parsimonious and structured mechanism to examine specifically how detachment or disengagement mediates the relationship between stress and well-being. Our study examines psychological detachment through SNS as a mediator and moderator between COVID-19 stressors and well-being.

According to Sonnentag and Fritz (2015), job stressors inhibit psychological detachment from work during non-work time—primarily because job stressors contribute to negative activation, a state that makes it difficult to psychologically detach from work. Thus, the model suggests that when employees are exposed to job stressors, they struggle to psychologically detach from work despite their need for detachment and recovery (De Croon et al., 2004). A lack of psychological detachment, in turn, further amplifies strain reactions and impairs affective states and well-being. Conversely, the stressor-detachment model proposes that psychological detachment attenuates the association between job stressors and strains (Sonnentag and Fritz, 2015). As a result, the stressor-detachment model suggests that the lack of psychological detachment is a partial mediator linking job stressors and strains.

The term detachment, which Etzion et al. (1998) introduced in the field of respite research, refers to ‘the individual's sense of being away from the work situation’ (p. 579). Thus, detachment denotes a subjective experience that goes beyond pure physical distance from one's workplace. To emphasise this experiential aspect, Sonnentag and Bayer (2005) introduced the concept of psychological detachment into research on stress and recovery. Specifically, psychological detachment from work during non-work time means psychologically disengaging oneself from work when away from the workplace (Sonnentag and Bayer, 2005). First, the concept of psychological detachment implies not being involved in work or work-related tasks. Second, it implies not thinking about work-related issues (Sonnentag and Fritz, 2007). Thus, psychological detachment from work during non-work time is a context-specific experience that may or may not occur when one is away from the workplace. The original conceptualisation of psychological detachment does not incorporate mentally disengaging and distancing oneself from work during work hours. Nevertheless, as Sonnentag and Fritz (2015) noted, it is plausible to assume that psychological (rather than physical) detachment can occur, for example, during lunch breaks.

Against this backdrop, we examine whether the stressor-detachment model can be extended to SNS use during a global pandemic. We define ‘psychological detachment through SNS use’ as temporarily mentally disengaging and refraining from activities or events that create stress (i.e. stressors) through SNS use. Therefore, psychological detachment through SNS use is an experience that helps individuals to avoid and recover from stressors associated with the COVID-19 pandemic.

3. Hypotheses development

Fig. 2 below summarises the research model and the hypotheses developed in the following subsections.

Fig. 2.

The research model.

3.1. Impact of stressors

The first stressor we adopted is work–family conflict due to COVID-19. Boundaries (e.g. physical, temporal and behavioural) structure and demarcate the roles that individuals maintain in various domains (Boswell and Olsen-Buchanan, 2007). Thus, work–family conflict can be conceptualised as a form of inter-role conflict, whereby the role demands of one domain interfere with the role demands of another domain (e.g. Boswell and Olsen-Buchanan, 2007; Greenhaus and Beutell, 1985; Kahn et al., 1964). IT use for remote work purposes typically affords greater work–family integration and thereby contributes to the blurring of boundaries between work and family domains (Batt and Valcour, 2003; Chesley et al., 2003; Fenner and Renn, 2004; Valcour and Hunter, 2005). Scholars have shown that technology-induced work–family conflict contributes to employees’ job stress (Yun et al., 2012). Therefore, it is plausible to consider work–family conflict as a stressor that ultimately creates psychological strain.

Due to the COVID-19 pandemic, the practice of working from home has increased exponentially across sectors. Combined with the widespread work-related use of smartphones (Yun et al., 2012), enterprise social networking (Mäntymäki and Riemer, 2016) and video conferencing (Kominers and Gonzalez, 2020), to name a few, absence from the physical workplace is likely to induce work–family conflict. By suddenly and forcefully blurring the line between the two life domains, the need to work from home can exacerbate conflicts between work and family demands (Eng et al., 2010; Golden et al., 2006; Tremblay et al., 2006; Voydanoff, 2005). In other words, the fact that family sounds and sights interrupt work meetings on Zoom and that such meetings consume time that one is expected to spend with family becomes a source of stress that can drive aversive states of strain.

Consequently, we posit that work–family conflict during COVID-19 can increase strain, which manifests in reduced psychological well-being. Indeed, research has linked work–family conflict to negative consequences, such as increased fatigue (Jansen et al., 2003), increased psychological strain (Kinnunen et al., 2006), increased depression and anxiety (Lapierre and Allen, 2006), reduced sleep quality (Williams et al., 2006) and reduced employee satisfaction (Hill, 2005), all which manifest in reduced well-being (Van der Doef and Maes, 1999). These associations are rooted in the transactional model of stress (Lazarus and Folkman, 1984) and the notion that well-being, which captures one's level of comfort, health and happiness, diminishes when stressors are introduced into one's life (Cooper, 2013). Consistent with such views, the stressor-detachment model likewise suggests that stressors have a direct impact on psychological well-being. IS research has also supported these relationships between stress and strain (see Whelan et al., 2020). Consequently, we propose the following hypothesis:

H1. Work–family conflict due to COVID-19 is negatively associated with psychological well-being.

Our second COVID-19-related stressor is perceived isolation. We draw on Marshall et al. (2007) and conceptualise perceived isolation as a psychological construct that captures one's perceptions of isolation from friends, peers and family members due to COVID-19. The absence of emotional and social interactions creates isolation perceptions. Here, we consider social isolation as a stressor because it induces an unpleasant and adverse state that people wish to avoid (Cacioppo and Cacioppo, 2014). Indeed, people are social animals with inherent needs for socialisation; if they are deprived of the ability to socialise, they react negatively (Dijksterhuis, 2005). In fact, prolonged social isolation can lead to depression (Cacioppo and Hawkley, 2009).

We argue that the measures taken to reduce the spread of COVID-19 have induced isolation. Restrictions regarding movement outside the home, the closure of restaurants and public spaces and the cancellation of various events and public gatherings have decreased opportunities for face-to-face interactions. In turn, this has forced people to rely increasingly on technology-mediated forms of communication and social interaction (Kominers and Gonzalez, 2020). While it is possible that technology-mediated interactions—for instance, via SNS or video calls—can partially compensate for the lack of face-to-face encounters, even the most advanced technology-mediated communications may not equal the information richness and social presence of face-to-face interactions (Yoo and Alavi, 2001; Waizenegger et al., 2020). The literature on technology-mediated communication has documented that electronic communication generally features lower levels of information richness and social presence than does face-to-face communication (Andres, 2002; Scott and Timmerman, 1999). Information richness refers to the extent to which a communication medium bridges different frames of reference, carries multiple cues, reduces equivocation and minimises ambiguity (Daft et al., 1987). Social presence, meanwhile, refers to the degree to which a communication medium is able to convey a sense of the communication parties’ physical presence, non-verbal signals and social cues (Daft et al., 1987). Therefore, despite the surge in technology-mediated communication and interaction during the COVID-19 pandemic, the lack of face-to-face social interactions has likely contributed to an increased sense of isolation.

As the transactional model of stress (Lazarus and Folkman, 1984) asserts, perceived isolation, like all stressors, can increase strain in the form of reduced well-being. In fact, several models support this assertion and show that social isolation is associated with higher levels of psychological strain and lower levels of job satisfaction (e.g. Bentley et al., 2016) and well-being (Shankar et al., 2015). Consequently, we propose the following hypothesis:

H2. Perceived isolation due to COVID-19 is negatively associated with psychological well-being.

The stressor-detachment model posits a relationship between stressors and psychological detachment. This relationship is rooted in the cognitive activation theory of stress (CATS; Ursin and Eriksen, 2004). According to CATS, in certain instances, individuals do not possess the necessary resources to manage or remove stressors. Because the COVID-19 pandemic is outside of their control, individuals may engage in passive responses, such as avoidant coping, to forget about or detach themselves from the stressors created by COVID-19. In the pandemic context, physical detachment (such as leaving one's house to work or see family) became largely infeasible. Therefore, we assume that most people turned to more feasible detachment vehicles, such as SNS use, to cope with their stressors. SNSs are effective tools of psychological detachment and escapism because they allow people to experience a state of flow and forget about their daily problems (Kırcaburun & Griffiths, 2019; Young et al., 2017). As such, we advance the following two hypotheses:

H3. Work–family conflict due to COVID-19 is positively associated with psychological detachment through SNS use.

H4. Perceived isolation due to COVID-19 is positively associated with psychological detachment through SNS use.

3.2. Impact of psychological detachment through SNS use

The stressor-detachment model (Sonnentag and Fritz, 2015) suggests that psychological detachment has a positive effect on psychological well-being. The rationale for this relationship is that stressors deplete an individual's resources and cause him or her to experience strain. To recover, people attempt to remove the stressors by detaching themselves from stressful events or situations, which, in turn, increases subjective well-being. However, the prior literature has reported inconsistent findings regarding the relationships between psychological detachment and positive outcomes. For example, prior studies have found both positive (e.g. Kühnel et al., 2009; Siltaloppi et al., 2009) and negative (e.g. Shimazu et al., 2012) relationships between psychological detachment and work engagement. Similarly, the prior literature has noted positive relationships between psychological detachment and life satisfaction (Fritz et al., 2010; Safstrom and Hartig, 2013), while also linking psychological detachment to vigour, joviality and serenity (Hahn et al., 2012). Nevertheless, because the majority of prior studies appear to support the view of the stressor-detachment model (see Sonnentag and Fritz, 2015 for a summary of the literature), we theorise a positive relationship between psychological well-being during the COVID-19 pandemic and psychological detachment through SNS use. Our logic is that detachment through SNS use (e.g. watching music videos and reading memes posted by friends) can energise people, allow them to forget about COVID-19 stressors and thereby improve their comfort and happiness, which manifests in heightened well-being. Thus, we advance the following hypothesis:

H5. Psychological detachment through SNS use is positively associated with psychological well-being.

The stressor-detachment model (Sonnentag and Fritz, 2015) also proposes that psychological detachment can moderate the effect of stressors on well-being. The possible moderating role of psychological detachment is rooted in the conservation of resources theory (CRT; Hobfoll, 1989), which describes the human motivation to maintain current resources and secure new ones. According to CRT, when humans lose their resources, they experience strain. However, humans tend to invest in new resources to protect against the loss of their current resources, to recover from such losses or to gain new resources. Applying this to the work–family conflict context, Moreno-Jiménez et al. (2009) argued that work–family conflict leads to strain because resources are lost during the attempt to balance the demands of work and family life. They also suggested that personal resources in the form of recovery experience (i.e. psychological detachment) are important to reduce the negative effects of stressors on well-being. Consequently, Moreno-Jiménez et al. (2009) proposed and empirically validated the existence of psychological detachment's moderating effect on the relationships between stressors (specifically, work–family conflict) and well-being. In a later study, Moreno-Jiménez et al. (2012) found that psychological detachment also attenuated the relationship between role conflict and anxiety. In a longitudinal study, Sonnentag et al. (2010) confirmed the moderating role of psychological detachment between stressors and strain, and in another study, Sonnentag et al. (2013) reported that psychological detachment attenuated the relationship between emotional conflict and decreased well-being. Using similar logic, we suggest that detachment through SNS use can help individuals to preserve their existing mental resources and regain new ones, thereby creating a buffer against or weakening the negative impact of stressors (in our case, isolation and work–family conflict) on well-being. Consequently, we propose the following two hypotheses:

H6. Psychological detachment through SNS use weakens the negative association between work–family conflict due to COVID-19 and psychological well-being (H1) such that this association becomes less negative when psychological detachment is higher.

H7. Psychological detachment through SNS use weakens the negative association between perceived isolation due to COVID-19 and psychological well-being (H2) such that this association becomes less negative when psychological detachment is higher.

4. Methodology

4.1. Data collection

We designed a survey questionnaire by adapting validated scales to our context. Psychological well-being, isolation and work–family conflict were measured on a five-point semantic scale ranging from ‘Never’ to ‘Always’ (Schriesheim and Schriesheim, 1978), while psychological detachment through SNS use was measured on a five-point Likert scale with ‘Strongly disagree’ and ‘Strongly agree’ as anchors. The amount of SNS use was measured on five-point (SNS use 1) and seven-point (SNS use 2) semantic scales. After drafting the initial survey questionnaire, we conducted a pilot test with ten respondents (Facebook users) and made minor adjustments based on their feedback. Table 2 presents the items for measuring the constructs and the sources from which they were adapted.

Table 2.

Research constructs and their measurements.

Construct		Measurement
Perceived isolation (Marshall et al., 2007)		Please evaluate your social circles during the COVID-19 pandemic and answer the following. Answer options: Never…Always
	Isolation 1	I have friends available to me.
	Isolation 2	I have one or more friends available with whom I talk about day-to-day problems.
	Isolation 3	I have friends available whom I can depend on when I have a problem.
	Isolation 4	I have enough people available with whom I can talk about things that matter to me.
Work–family conflict (Netemeyer et al., 1996)		Please comment on the following statements based on your experiences during the COVID-19 pandemic. Answer options: Never…Always
	Work–Family 1	The demands of my work have interfered with my family life.
	Work–Family 2	The amount of time my job takes up has made it difficult to fulfil family responsibilities.
	Work–Family 3	Things I have wanted to do at home have not been done because of the demands my job puts on me.
	Work–Family 4	My job has produced a strain that has made it difficult to make changes to my plans for family activities.
	Work–Family 5	Due to work-related duties, I have had to make changes to my plans for family activities.
Psychological detachment through SNS use (Sonnentag and Fritz, 2007)		Please evaluate your Facebook use during the COVID-19 pandemic and comment on the following statements. Answer options: Strongly disagree…Strongly agree
	Detachment 1	By using Facebook, I forget about COVID-19.
	Detachment 2	By using Facebook, I don't think about COVID-19 at all.
	Detachment 3	By using Facebook, I distance myself from COVID-19.
	Detachment 4	By using Facebook, I get a break from thinking about COVID-19.
Psychological well-being (World Health Organisation WHO, 1998)		Please answer the following questions based on your experiences during the COVID-19 pandemic. Answer options: Never…Always.
	Well-Being 1	I feel cheerful and in good spirits.
	Well-Being 2	I feel calm and relaxed.
	Well-Being 3	I feel active and vigorous.
	Well-Being 4	I wake up feeling fresh and rested.
	Well-Being 5	My daily life is filled with things that interest me.
Amount of SNS use	SNS use 1	How many times do/did you visit Facebook a day during the COVID-19 pandemic?
		Answer options: Less than once, Once or twice, 3–5, 6–10, More than 10
	SNS use 2	How much time do/did you spend on Facebook a day (all devices included) during the COVID-19 pandemic?
		Answer options: Less than 5 min, 5–15 min, 16–30 min, 30 min–1 hour, 1–2 h, 2–3 h, More than 3 h

We collected the data in early May 2020 when many countries were operating under COVID-19 related restrictions (lockdown). We collected data using Amazon Mechanical Turk (MTurk), which prior research has found to be a useful platform for survey data collection (Cheung et al., 2017; Hauser and Schwarz, 2016; Walter et al., 2019). Compared to student or community samples, MTurk provides a unique opportunity to obtain data with global coverage of the working population.

We took five steps to maximise the quality of our sampling procedure. First, we used ‘Facebook account holder’ as a sampling criterion in MTurk. Second, we offered the respondents 1.5 USD compensation for completing the survey. Third, we asked the respondents whether they had used Facebook during the COVID-19 pandemic, been in lockdown and worked from home during the pandemic. Fourth, we included several screening questions (e.g. Please select ‘agree’ in response to this statement) in the survey. Fifth, we prevented multiple responses from individual respondents by using MTurk ID as an identifier and informing the respondents that only one response qualified for compensation.

We obtained a total of 510 responses. Forty-eight respondents failed to answer the screening questions correctly or had not been in lockdown during the COVID-19 pandemic. Therefore, we removed these respondents from the sample. Finally, we removed responses from students, stay-at-home parents, retirees and respondents who had not worked from home at all during the COVID-19 pandemic. After these omissions, the dataset to be analysed included 398 responses. Table 3 presents the respondents’ demographics.

Table 3.

Background information of respondents.

Country of residence		Sex		Age		Household with children (Age below 13)
USA	46.3%	Female	39.7%	20–29	30.9%	Yes	49.7%
India	39.0%	Male	60.3%	30–39	40.7%	No	50.3%
Brazil	3.7%			40–49	17.3%
Canada	2.7%			50–59	7.0%
UK	1.0%			60–69	4%
Other	7.3%

As Table 3 shows, 60% of the respondents were male, and approximately 70% of the respondents were below 40 years of age. Approximately half of the respondents had one or more children younger than 13 years of age living in the same household. In terms of location, most of the respondents (46.3%) were from the USA, followed by India with 39%. Approximately 70% of the respondents reported they had been in total or almost total lockdown during the pandemic. Ca. 57% of the respondents had been in lockdown for more than two months, and 34% had lived under lockdown for one to two months.

4.2. Data analysis and results

We used AMOS 26 to analyse the data. As a first step, we tested the measurement's convergent validity. Table 4 indicates that all item loadings exceeded 0.7, average variance extracted values (AVEs) exceeded 0.5 and composite reliability (CR) values exceeded 0.8, except for the amount of SNS use, which was marginally below the 0.8 threshold (0.796). Thus, we concluded that the measurement exhibited solid convergent validity.

Table 5.

Discriminant validity.

	Isolation	Work–family conflict	Well-being	Psych. detachment	Age	Sex	Children in household	Lockdown length	Amount of SNS use	Lockdown degree
Isolation	0.856
Work–family Conflict	n.s.	0.892
Well-being	−0.521***	n.s.	0.859
Psychological detachment	0.128*	0.347***	0.232***	0.861
Age	n.s.	−0.308***	n.s.	n.s.	n/a
Sex	n.s.	n.s.	n.s.	n.s.	0.170***	n/a
Children in household	n.s.	0.332***	0.124*	0.253***	−0.139**	n.s.	n/a
Lockdown length	n.s.	n.s.	−0.107*	−0.168**	n.s.	n.s.	n.s.	n/a
Amount of SNS use	0.161**	0.328***	n.s.	0.248***	n.s.	n.s.	0.227***	n.s.	0.754
Lockdown degree	n.s.	n.s.	n.s.	−0.232***	n.s.	n.s.	−0.122*	0.271***	n.s.	n/a

Table 4.

Construct validity.

	Item	Mean	SD	Loading	CR	AVE
Isolation	Isolation 1 (R)*	3.595	0.986	0.861	0.901	0.735
	Isolation 2 (R)*	3.543	1.025	0.823
	Isolation 3 (R)*	3.595	1.072	0.871
	Isolation 4 (R)*	3.706	1.039	0.774
Psychological well-being	Well-being 1	3.354	0.919	0.802	0.877	0.746
	Well-being 2	3.394	0.935	0.748
	Well-being 3	3.216	1.013	0.741
	Well-being 4	3.224	1.052	0.788
	Well-being 5	3.389	1.012	0.755
Psychological detachment through SNS use	Detachment 1	2.972	1.736	0.877	0.910	0.741
	Detachment 2	2.892	1.803	0.869
	Detachment 3	3.595	2.000	0.802
	Detachment 4	3.530	1.871	0.836
Work–family conflict	Work–family 1	2.721	1.202	0.896	0.951	0.799
	Work–family 2	2.651	1.230	0.905
	Work–family 3	2.714	1.237	0.879
	Work–family 4	2.628	1.259	0.885
	Work–family 5	2.741	1.240	0.896
Amount of SNS use	SNS use 1	3.324	1.224	0.745	0.796	0.570
	SNS use 2	4.093	1.658	0.877

*(R) Reversed.

Next, we ensured our measurement's discriminant validity. To this end, we compared the inter-construct correlations with the square roots of the AVEs (see Table 5). As Table 5 shows, the inter-construct correlations fell below the diagonally presented square roots of the AVEs. Consequently, we concluded that the constructs exhibited good convergent and discriminant validity and proceeded to evaluate the model fit. In brief, the measurement model exhibited good fit with the data (χ²/DF = 1.570; GFI = 0.929, AGFI = 0.901, NFI = 0.935, CFI = 0.975, SRMR = 0.034 and RMSEA = 0.038).

Finally, we tested for common method bias (CMB) and multicollinearity. We used the common latent factor approach (Podsakoff et al., 2003) to test for CMB. To this end, we included a common latent method factor in the measurement model. All measurement items loaded higher on their intended constructs than on the common method factor, suggesting that CMB was not a significant concern. We tested for multicollinearity using SPSS. The highest variance inflation factor (VIF) value was 1.143, indicating that multicollinearity did not distort the results.

After ensuring the validity of the measurement, we continued to evaluate the proposed structural model (see Fig. 3 ). To ensure the validity of the hypotheses testing, we included several control variables. To this end, we placed the respondents’ amount of SNS use, age, sex, lockdown length, lockdown degree and presence in the household of children below 13 years (binary variable) as predictors of psychological detachment through SNS use and psychological well-being.

Fig. 3.

Structural model results.

The results from testing the structural model provided empirical support for three (H1, H3 and H5) out of seven of the hypothesised relationships. In addition, the data provided partial support for one hypothesis (H6). Both the evidence supporting and refuting the hypotheses was informative. Work–family conflict did not have a significant negative effect on psychological well-being. Thus, H1 was not supported. Meanwhile, H2 was supported because perceived isolation exerted a negative effect on psychological well-being. In addition, work–family conflict had a significant positive effect on psychological detachment. Thus, H3 received support. However, perceived isolation did not have any statistically significant effect on psychological detachment through SNS use. Hence, H4 was not supported. Consistent with the stressor-detachment model, psychological detachment through SNS use had a positive effect on psychological well-being. Hence, H5 was supported.

We found a weak positive interaction effect of psychological detachment through SNS use and work–family conflict on psychological well-being. This was an interesting observation because the effect of work–life conflict on psychological well-being was not significant. To investigate the issue further, we divided the data into two subsamples with below average and above average psychological detachment through SNS use. The results indicated that the effect of work–life conflict was statistically non-significant in the low detachment subsample but positive (0.099*) in the high detachment subsample. This suggests that the effect of work–life conflict becomes positive amongst users with higher levels of psychological detachment through SNS use. This observation deviates from the stressor-detachment model, which posits that psychological detachment, in general, attenuates the negative impact of stressors on well-being. However, our results nevertheless indicate that psychological detachment can interact with stressors in a way that reinforces well-being. Thus, we conclude that H6 received partially support. Finally, we did not find any significant effect for psychological detachment through SNS use on the magnitude of the negative association between perceived isolation and psychological well-being. Thus, H7 was not supported. Altogether, the model explained 35.9% of the variance in psychological well-being and 22.0% of the variance in psychological detachment through SNS use. amongst the control variables that we employed, the presence of children in the household and the amount of SNS use had statistically significant, albeit weak, effects on psychological well-being. The effect of the presence of children in the household had a positive effect on well-being, while the amount of SNS use had a negative effect. In addition, both the degree of lockdown and the length of lockdown had negative effects on psychological detachment through SNS use. This means that the greater extent to which and the longer people were confined to their homes due to COVID-19, the less they relied on SNS for psychological detachment. One way to interpret these results is that the attractiveness of SNS as a coping tool may depend on interactions that occur in person. In other words, individuals may only view SNS use as a viable coping tool when it supports face-to-face relationships and not when it functions as a standalone tool for online interaction. It might be interesting to examine this issue in future research. Finally, the amount of SNS use had a positive effect on psychological detachment through SNS use. This result is intuitive because people who are generally more active on SNS presumably also have more opportunities for psychological detachment through SNS use. Finally, gender and age did not have any significant effect on psychological well-being or psychological detachment through SNS use.

5. Discussion

5.1. Key findings

Our empirical results both confirm and contrast with our expectations. We hence leverage both the significant and non-significant findings to generate important insights. First, consistent with the stressor-detachment model (Sonnentag and Fritz, 2015) and the theories underlying it (e.g. Ursin & Eriksen, 2010; Ganster and Rosen, 2013), we identify perceived isolation as a stressor that increases strain in the form of reduced psychological well-being. However, in contrast with the stressor-detachment model, our research does not reveal work–life conflict to significantly decrease psychological well-being in the whole sample.

Second, our results show that work–family conflict motivates psychological detachment through SNS use, while perceived isolation does not. This observation deviates from the stressor-detachment model, which posits that all stressors are likely to motivate psychological detachment (Sonnentag and Fritz, 2015). Our results thus suggest that not all stressors are alike in terms of psychological detachment. This can be explained post-hoc by the theory of learned helplessness (Maier and Seligman, 1976), which suggests that some stressors—indicating helplessness and the lack of opportunity—override individuals’ motivation and reduce their engagement in coping behaviours. In our context, it is reasonable to assume that people who had many SNS contacts with whom they could interact regularly felt less isolated during the pandemic. In this case, SNS can be viewed as an effective means for coping with isolation. In contrast, people who did not have a strong SNS network of friends before the pandemic were more likely to feel isolated (having limited means to form new relationships during the lockdown) and consequently more likely to view SNS as a less effective means of dealing with isolation. Consistent with helplessness theory (Maier and Seligman, 1976), socially isolated people are more likely than others to consider the use of SNS to be futile and thus prefer other means of coping. We call for future research to delve deeper into and directly examine this explanation.

The non-significant association between perceived isolation and psychological detachment is also supported by motivation–opportunity theories (Fazio and Towles-Schwen, 1999; Olson and Fazio, 2008), which posit that actions are based on motivation and opportunities. While people may be motivated to cope with isolation, SNS may not provide the best opportunity to do so. Logically, people with a limited circle of friends who can provide social support through SNS are likely to feel more isolated. Conversely, people with a large circle of friends who provide social support through SNS are less likely to feel socially isolated. Together, our findings suggest that seeking to disengage via SNS does not appear to be a feasible solution for individuals experiencing high levels of isolation due to COVID-19. This is not to say, however, that SNS use cannot offer an invigorating break from everyday problems; rather, this solution simply has a limited role in helping individuals to cope with more serious conditions, such as social isolation due to COVID-19.

Third, aligned with the stressor-detachment model (Sonnentag and Fritz, 2015), we demonstrate the positive impact of psychological detachment through SNS use on individuals’ psychological well-being. This implies that people did turn to SNSs to detach themselves from COVID-19 stressors and thereby reinforce their psychological well-being.

Fourth, our results indicate that the effect of work–family conflict on psychological well-being is contingent on an individual's level of psychological detachment through SNS use. Interestingly, while the effect of work–family conflict on psychological well-being was not significant in the whole sample, work–family conflict did have a positive impact on psychological well-being amongst users experiencing higher degrees of psychological detachment. This finding adds to the extant research on the stressor-detachment model (Sonnentag and Fritz, 2015) by empirically demonstrating a different way in which psychological detachment can affect the relationship between stressors and strain.

5.2. Theoretical implications

First, our findings advance the emerging research on the psychosocial implications of global pandemics (e.g. Islam et al., 2022; Nanath et al., 2022; Khan, 2021; Laato et al., 2020; Rai, 2020) as well as the research on the psychosocial implications of social media (Tandon et al., 2020, 2021; A. Tandon, Dhir and Mäntymäki, 2021, 2021; Tandon et al., 2022; Mäntymäki and Islam 2016; Islam et al., 2019) In particular, we contribute to research regarding the role of psychological detachment (Sonnentag and Bayer, 2005; Sonnentag and Fritz, 2015) in coping with stressors. To this end, we extend the application of psychological detachment to the IT domain. Based on our empirical results, we suggest that the short-term detachment from a major long-term stressor, such as the COVID-19 pandemic, that SNS use affords can exert a positive impact on psychological well-being. We further extend research on psychological detachment (Sonnentag and Bayer, 2005; Sonnentag and Fritz, 2015) by integrating this model with helplessness and motivation–opportunity theories (Fazio and Towles-Schwen, 1999; Maier and Seligman, 1976). This integrated perspective demonstrates that not all stressors equally motivate psychological detachment. As our results indicate, work–family conflict drives psychological detachment through SNS use, whereas perceived isolation does not. This suggests that people choose coping strategies by evaluating the fit between such strategies and the relevant stressors. This notion paves the way for studies investigating not only IT but other coping strategies. From a broader perspective, our findings imply that psychological detachment through SNS use mediates the relationship between stressors and well-being, as suggested by Sonnentag and Fritz (2015). However, the specific nature of this mediation is contingent upon the type of stress.

Second, our study contributes to the literature on IT-mediated strategies for coping with stressors (Beaudry and Pinsonneault, 2005; Stein et al., 2015; Tarafdar et al., 2020) by theorising and empirically demonstrating that the use of hedonic IS, such as SNS, can help individuals to cope with stressful events and situations that are essentially not IT related or induced. In doing so, we reveal another side of technology and stress, according to which IT can be an effective means of coping with some stressors. Consistent with our findings, we call for future research to take a more balanced approach to and perhaps an integrative perspective on IT and stress. Future scholars could, for instance, examine the simultaneous roles of IT as a stressor and as a means of alleviating non-IT stress. Collectively, our findings add to the literature on technology-mediated strategies for coping with stressors (e.g. D'Arcy et al., 2014; Tarafdar et al., 2020).

Third, with respect to types of stress, we find that psychological detachment through SNS use moderates the relationship between work–family conflict and psychological well-being and that the effect of work–family conflict on psychological well-being during the COVID-19 pandemic was, in fact, positive amongst users experiencing higher degrees of psychological detachment through SNS use. This may imply that under exceptional circumstances, such as the COVID-19 pandemic, work–family conflict represents positive stress (eustress) that supports well-being, while perceived isolation represents negative stress (distress), which has a detrimental effect on well-being (Folkman, 1997; Nelson and Simmons, 2003). However, this notion merits further research.

Fourth and finally, regarding our theoretical foundation—that is, the stressor-detachment model (Sonnentag and Fritz, 2015), our data partially support the moderating effects theorised therein. These observations appear to echo the findings of prior empirical studies, which have found that psychological detachment does not moderate all relationships between stressors and strain (see Sonnentag and Fritz, 2015). For example, Moreno-Jiménez et al. (2012) reported that psychological detachment did not moderate the relationship between conflict and somatic complaints. Furthermore, Safstrom and Hartig (2013) demonstrated the non-significant moderating role of psychological detachment. In contrast, Moreno-Jiménez et al. (2009) found that psychological detachment attenuated the relationship between work–family conflict and well-being. Because we adapted the stressor-detachment model to the context of the COVID-19 pandemic, which is an exceptional situation, and contextualised the domain of psychological detachment to SNS use, additional empirical research in other contexts is required to verify and extend our observations. Table 6 below summarises the current study's main contributions.

Table 6.

Key contributions.

Source of contribution	Description	Area of contribution
Under-researched but highly societally relevant phenomenon regarding stress and impaired psychological well-being due to the COVID-19 pandemic (Corley and Gioia, 2011; Hambrick, 2007; Weber, 2012)	Examination of two highly relevant stressors—perceived isolation and work–family conflict—and their impact on psychological well-being during the peak of the COVID-19 pandemic	Psychosocial implications of global disruptive events, such as pandemics (e.g. Islam et al., 2022; Laato et al., 2020; Rai, 2020). Dark side of social media (e.g. Tandon et al., 2020, 2021; A. Tandon, Dhir and Mäntymäki, 2021, 2021; Tandon et al., 2022; Mäntymäki and Islam 2016; Islam et al., 2019)
New constructs (Whetten, 1989)	Introduction of a new construct—psychological detachment—through SNS use	IT-mediated coping with stressors (Tafardar et al., 2020; D'Arcy et al., 2014) Contextualisation of the stressor-detachment model to IS (Sonnentag and Fritz, 2015)
Novel associations between constructs (Whetten, 1989)	Empirical evaluation of the direct, mediating and moderating effects of psychological detachment through SNS use on psychological well-being, which demonstrated the positive impact of psychological detachment through SNS use on psychological well-being, the positive interaction effect of work–family conflict and psychological detachment through SNS use on psychological well-being and the ineffectiveness of detachment-orientated SNS use as a strategy to cope with perceived isolation	IT-mediated coping with stressors (Tafardar et al., 2020; Schmalz et al., 2015; Pahayahay and Khalili-Mahani, 2020)

5.3. Implications for policy and practice

Our findings provide insights into the psychosocial implications of major disruptive events, such as global pandemics. In particular, the strong negative relationship between perceived isolation and psychological well-being we observed underscores the widely documented detrimental effects of the COVID-19 pandemic on various aspects of physical and mental health and well-being (see, for example, Hwang et al., 2020; Pietromonaco and Overall, 2022). This implies that fighting pandemics through measures that induce isolation may harm people's mental health and well-being. Additional research is thus required to develop a more deliberate understanding of these evident trade-offs.

Second, the observation that psychological detachment through SNS use is not a preferred strategy for coping with isolation further highlights the importance of new interventions for addressing isolation and promoting psychological well-being. Pandemics, such as COVID-19, pose a threat not only to people's physical health but also to their psychological health. Governments and nongovernmental organisations around the world have undertaken various measures to mitigate the pandemic's physical effects. However, they have devoted less effort to promoting psychological well-being, particularly by addressing the isolation caused by measures intended to combat the pandemic (Islam and Islam, 2020). SNSs may help, for example, by introducing features that enable individuals to meet new friends during lockdown. Such efforts may be especially important because relying on existing networks does not seem to suffice as a coping strategy for those who already feel isolated. Health professionals can also utilise SNSs to deliver more formal treatments to those who feel isolated. Ultimately, it seems that SNS can and should be used more and more effectively to fight isolation.

Third work–family conflict has generally been considered a source of negative stress; however, under exceptional circumstances, such as pandemics, which have implications for people's health and livelihoods, work–family conflict can also represent positive stress, which indirectly improves psychological well-being. Moreover, the positive relationship between work–family conflict and psychological detachment through SNS use implies that the stress caused by work–family conflict may motivate people to mentally disengage from the multiple threats caused by COVID-19. Thus, at least during the pandemic, employers should consider other metrics beyond work–family conflict when assessing the impact on employees who are working remotely.

6. Limitations and future research

The current study has several limitations that may indicate paths for future research. The main limitations can be attributed to the choice of research variables and the adopted research design. First, with respect to the choice of research variables, we included only two COVID-19 stressors in our theoretical model. While the extant research supports these choices, other stressors exist. For example, COVID-19 has arguably evoked various threats, including threats to physical health, potential unemployment and financial problems. Future studies should, therefore, extend our model by including additional stressors.

Second and likewise pertaining to the choice of variables, we employed psychological detachment via Facebook as a strategy for reducing the impact of stressors on psychological well-being. However, in addition to using Facebook or any other social media platform, people may attempt to detach themselves from COVID-19 stressors by using other technologies, such as video games, e-commerce and video-streaming services. Therefore, future research should investigate the extent to which people can cope with the stressors of COVID-19 by using these or other technologies.

Third, we employed psychological well-being as the sole dependant variable in our model. However, scholars can examine other important variables as consequences of psychological detachment through SNS use. Hence, future research that builds on the stressor-detachment model might include emotional exhaustion, depression and burnout, amongst other variables.

Fourth, our study utilised a cross-sectional research design. Thus, it offers only a snapshot of the rapidly changing pandemic situation. Consequently, additional research must employ longitudinal research designs.

Fifth and finally, people may vary in their ability to detach themselves from stressors. Furthermore, their ability to detach themselves might also depend upon contextual factors, such as the length of the lockdown and the number of children in the household, amongst others. These factors, in turn, may moderate the relationships in our proposed model. Therefore, future research should investigate the moderating effects of personality-related and contextual factors.

CRediT authorship contribution statement

Matti Mäntymäki: Conceptualization, Data curation, Formal analysis, Validation, Writing – original draft, Writing – review & editing. A.K.M. Najmul Islam: Conceptualization, Data curation, Formal analysis, Validation, Writing – original draft, Writing – review & editing. Ofir Turel: Formal analysis, Validation, Writing – original draft, Writing – review & editing. Amandeep Dhir: Writing – original draft, Writing – review & editing.

Biographies

Matti Mäntymäki is an Associate Professor of Information Systems Science at Turku School of Economics, University of Turku, Finland. His-workhas been published in outlets such as Information Systems Journal, Journal of Systems and Software, Information Technology & People, Information Systems Frontiers, Computers in Industry, Computers in Human Behaviour, International Journal of Hospitality Management, Technological Forecasting and Social Change, International Journal of Information Management, Journal of Medical Internet Research, Journal of Business Research, Internet Research, Journal of Retailing and Consumer Service, Electronic Markets, Behaviour & Information Technology, Enterprise Information Systems, and Communications of the Association for Information Systems, amongst others.

A.K.M. Najmul Islam is an Associate Professor at the LUT School of Engineering Science, LUT University, Finland. He also works as University Research Fellow at the Department of Future Technologies, University of Turku, Finland. Dr. Islam holds a PhD (Information Systems) from the University of Turku, Finland and an M.Sc. (Eng.) from Tampere University of Technology, Finland. His research has been published in top outlets such as IEEE Access, Information Systems Journal, Journal of Strategic Information Systems, European Journal of Information Systems, Technological Forecasting and Social Change, Computers in Human behaviour, Journal of Medical Internet Research, Internet Research, Computers & Education, Information Technology & People, Telematics & Informatics, Journal of Retailing and Consumer Services, Communications of the AIS, and Behaviour & Information Technology, amongst others.

Ofir Turel is a Professor of Management Information Systems in the Mihaylo College of Business and Economics at California State University Fullerton , and a Scholar in Residence at the Decision Neuroscience Program, Department of Psychology at the University of Southern California (USC). Dr. Turel's research interests include a broad range of behavioral, bio-physiological, and managerial issues in various information systems use contexts. He has published more than 90 articles in business, research methods, psychology, psychiatry, and medicine journals. Examples of business outlets include MIS Quarterly, Journal of MIS, MIT Sloan Management Review, European Journal of Information Systems, Information Systems Journal, and Communications of the ACM. Examples of psychology outlets include Journal of Psychiatric Research, Cognitive Affective and Behavioral Neuroscience, Substance Use and Misuse, Social Neuroscience, Progress in Neuropsychopharmacology & Biological Psychiatry, Scientific Reports, andPsychiatry Research: Neuroimaging. .

Amandeep Dhir (DSc, PhD) is a Professor of Research Methods at University of Agder, Norway. He is also a visiting professor at Norwegian School of Hotel Management, University of Stavanger, Norway. His-research appears in the Journal of Retailing and Consumer Services, Journal of Destination Marketing & Management, International Journal of Contemporary Hospitality Management, International Journal of Hospitality Management, Technological Forecasting and Social Change, International Journal of Information Management, Computers in Human Behaviour, Computers in Industry, Journal of Cleaner Production, Food quality and preferences, Appetite, Information Technology & People, Australasian Marketing Journal, Enterprise Information Systems amongst others