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

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.


Introduction
The COVID-19 pandemic dramatically decreased people's face-toface 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  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.

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 1 Facebook usage is surging, but the company warns this surge may be temporary.
https://www.theverge.com/2020/4/29/21241845/facebook-q1-2020-earnings-coronavirus-covid-19-daily-users-engagement-up 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.
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.

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 stressordetachment 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.
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. 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;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.

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 2 COVID-19 is making America's loneliness epidemic even worse, Time Magazine, 8 May 2020 https://time.com/5833681/loneliness-covid-19/ 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 wellbeing. 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 Fig. 1. Stressor-detachment model (Sonnentag and Fritz, 2015).  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.

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.

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.
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.
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.

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.
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 (χ 2 /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)   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.
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.

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 stressordetachment 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  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.

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(Tandon et al., , 2021Mäntymäki, 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.

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 . 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.  (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., 2020Tandon et al., , 2021Mäntymäki, 2021, 2021;Tandon et al., 2022;Mäntymäki and Islam 2016; 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)

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 videostreaming 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.