Sleeping to Support: An Examination of the Relationship between Leader Sleep and Positive Support Behaviours
The unhealthful notion that great leaders do not get enough sleep is perpetuated by the competitive workplace atmosphere of today. Many first-person testimonies from well-known leaders, such Bill Gates and Margaret Thatcher, who have acknowledged to previously forgoing sleep in order to gain an advantage as a leader or to advance their responsibilities, reflect this (Gates, 2019; Lashbrooke, 2020). Similar actions have also been taken by previous American presidents. For instance, Donald Trump has been cited as claiming that he "never sleeps and that people who sleep are lethargic" and Barack Obama was well-known for working instead of resting when he was president (Berger, 2018; Shear, 2016). (Le, 2019; Smith, 2017). Moreover, Vice President Kamala Harris said in a precampaign interview that she does not get "quite enough" sleep, while the current President of the United States, Joe Biden, is known for dozing off during President Obama's debt address in 2011. (New York Times, 2019). In fact, practically all of the Democratic contenders for president in 2020 indicated that they don't get enough sleep in the same interview (New York Times, 2019).
These stories are backed by organisational research, which indicates that people think sleeping less is associated to professional success. Participants thus believed effective leaders slept less than the typical worker (Svetieva et al., 2017). According to other studies (e.g., Kerstedt et al., 2004; Jackson et al., 2013; Luckhaupt et al., 2010; Svetieva et al., 2017; Ursin et al., 2009;), supervisor-level employees experience the shortest sleep durations and the most fatigue. This indicates that unhealthful sleep beliefs and practises are to blame.
Leaders in the workplace reinforce attitudes. A study by the United States Bureau of Labor Statistics found that paid work time was the main waking activity substituted for sleep (Basner et al.2012). This larger social tendency is mirrored in this( poll.al, 2014). Particularly, bosses are more prone to lengthen working hours at the expense of their personal time (such as sleep) (Babbar & Aspelin, 1998; Ruderman et al., 2017). When considered together, these research point to a national issue about leaders and chronic sleep deprivation (i.e., consistently obtaining less than ideal amounts of sleep).
The American Academy of Sleep Medicine, the Sleep Research Society, and the National Sleep Foundation advise a minimum of 7 hours of sleep each night for adults on a regular basis, as well as high levels of quality sleep (Hirshkowitz et al., 2015; Ohayon et al., 2017; Watson et al., 2015).
Approximately 83.6 million US people, or more than one third of Americans, were found in a recent research by the Centers for Disease Control and Prevention to consistently go below the necessary sleep guidelines (Liu et al., 2016). Given its ubiquity and accompanying effects, sleep deprivation may be highly harmful for the individual, the company, and society. Sleep has been linked to a number of health outcomes, including depressive, anxiety, depersonalization, and emotional tiredness. It has also been linked to a number of organisational outcomes, including job satisfaction, engagement, performance, safety, and absenteeism (e.g., Barnes & Watson, 2019; Litwiller et al., 2017). As a result, insufficient sleep causes roughly 1.2 million working days to be missed each year in the United States (Hafner, 2017; Shockey & Wheaton, 2017). Restricted sleep may also have a significant financial effect. For instance, National estimates of the economic impact of sleep deprivation approach $411 billion in medical and employment-related costs per year, or 2.3% of the US gross domestic product (Hafner, 2017; Kiley et al., 2019). Unfortunately, little study has been done on the possible effects of sleep deprivation on the population of business leaders.
Reviews and meta-analyses on sleep in the workplace reveal the restricted emphasis on regular employees as opposed to leaders in the industry (Khubchandani & Price, 2020). For instance, two recent meta-analyses ignored level within the company and experimentally investigated the effect of sleep on job outcomes and work performance for general workers (Henderson & Horan, 2021; Litwiller et al., 2017). A comprehensive assessment of the longitudinal intervention literature on psychosocial stresses and sleep quality among working people, but not necessarily leaders, was also carried out by Van Laethem and colleagues (2013). Barnes (2012) examines and combines sleep and self-regulation studies, concentrating on employee sleep and job results while ignoring organisational level. Another example is a meta-analysis that looked at sleep issues and safety results in the overall working population (Uehli et al., 2014). Researchers have urged for further study on the connection between sleep and performance among leaders in the workplace, notwithstanding the importance of knowing the association between sleep and job results among ordinary workers (e.g., Gaultney, 2014; Rogers et al., 2019). Due to their positions within the company, leaders' attitudes, actions, and choices are likely to have significant effects and may have an influence on a large number of individuals (Ruderman et al., 2017; Kaluza et al., 2021). Since leaders may be more prone to sleep loss, it is crucial to take into account how sleep is related to their capacity to do the specific activities associated with their position. Leaders who get enough sleep are essential to the general performance of their teams and businesses.
Individual research has recently begun to look at the connection between sleep and different leadership outcomes. The method through which leaders influence their followers is referred to as leadership (Antonakis & Day, 2018; Bass & Bass, 2008; Schonfeld & Chang, 2017; Truxillo et al., 2015). For instance, articulating a vision, securing collaboration, and modifying the attitudes and behaviours of followers are all examples of leadership.
Researchers contend that inconsistent sleep quality (i.e., feeling rested and having trouble falling and staying asleep) might be linked to abusive leadership (i.e., hostile verbal and nonverbal conduct) (Barnes et al., 2015; Tariq et al., 2019). Lack of sleep also heightens a leader's propensity to disregard and evade obligations related to their leadership job (Olsen et al., 2016). But the majority of this body of study focuses on the possible drawbacks of leader sleep for organisational performance. There are a few studies that investigate the relationship between leader sleep and subsequent effective leadership philosophies like charismatic leadership (leaders who captivate and persuade followers that their group's mission is extraordinary; Conger et al., 2000) and transformational leadership (i.e., leaders who encourage and empower followers to grow and achieve individual and collective goals; Barnes et al., 2016; Bass & Riggio, 2010; Byrne et al., 2014; Olsen et al., 2016). Although organisational research is starting to look at sleep as a significant aspect in leadership, little is understood about the function of sleep in a leader's effectiveness.
The capacity to exhibit constructive conduct at work rather than just the absence of bad behaviour. In light of this, it's critical to comprehend the connections between leader support, a closely similar but separate concept, and leader sleep. Support is one of the numerous actions that a successful, high-caliber leader must do, although it is rarely included in general definitions of leadership (e.g., van Dam & van der Helm, 2016). The relationship between supervisor support and sleep is now being studied in a similar area of study. Studies have specifically shown that leader support may enhance staff sleep (e.g., Berkman et al., 2010; Crain et al., 2014; Sianoja et al., 2020). Although significant, this line of study misses the connection between a leader's personal sleep and their capacity to exhibit helpful, supporting actions. Examining leader support is crucial since prior research indicates that it is particularly advantageous for both workers and the firm (e.g., Hammer et al., 2009; Hammer et al., 2013; Kelloway et al., 2017; Koch & Binneweis, 2015; Las Heras et al., 2015). But the causes of these effective leadership practises have largely been ignored (e.g., Crain & Stevens, 2018; Byrne et al., 2014). This is a fundamental challenge for the atmosphere and productivity of the workplace: How can managers support and assure the wellbeing of their staff members if they are themselves experiencing the negative effects of sleep deprivation? It is crucial for academics and practitioners to comprehend how to foster and sustain these supporting behaviours among leaders due to the well-established advantages of these behaviours.
Although leaders are essential for enhancing organisational and employee-level results, prior study has neglected to take into account signs of supportive leader behaviour.
Behaviours. In particular, there are three types of support behaviours that are particularly pertinent to this study: general supervisor support (i.e., the leader expressing care and concern for their employees or directly assisting them; House, 1981; Kossek et al., 2011), family-supportive supervisor behaviours (FSSB), and sleep leadership (i.e., actions that support employees in getting enough sleep). The general supervisor support is all-encompassing and concentrates on support within the workplace, the FSSB is made up of support behaviours for nonwork demands, and sleep leadership refers to support for employee sleep health. Each of these positive leader support behaviours targets a different area of the lives of the employees. By investigating leader sleep as a precursor to a deliberately selected collection of distinctive but significant positive leader support behaviours in the workplace; this research seeks to further this discourse.
Few studies have investigated probable mechanisms by why this association arises in the few research that has evaluated the connection between leader sleep and subsequent conduct. The theory that has received the greatest support in prior studies contends that resources are the processes linking sleep to subsequent results (e.g., Barnes, 2012; Barnes et al., 2015; Baumeister, 2003). However, resource-based procedures have been criticised for being overly general (e.g., "almost everything beneficial may be deemed a resource"; Halbesleben et al., 2014, p. 1337) and for impeding the advancement of research and practise (e.g., Friese et al., 2019; Thompson & Cooper, 2001). Research is thus required to pinpoint more precise resources that might be involved in the link between sleep and leadership style. The current research specifically examines emotional weariness as a significant explanatory factor. The basic component of burnout is emotional exhaustion, which is a particularly persistent, affective type of work-related stress that may manifest as weariness, job-related melancholy, psychosomatic problems, and anxiety (Demerouti et al., 2001; Gaines & Jermier, 1983; Wright & Cropanzano, 1998). Understanding the influence emotional weariness has on key organisational outcomes including job performance, organisational citizenship behaviours, and turnover intentions is crucial in the workplace (e.g., Cropanzano et al., 2003). Beyond these primary effects, little is known about how emotional tiredness may affect particular aspects of performance, such leaders' supportive actions.
According to humanities assignment help, current research adds three new theoretical perspectives to the existing literature on organisational health. First, little study has been done on the connection between sleep and a leader's capacity to exhibit supportive behaviours.
It's essential to know how to curb bad leadership habits at work, but concentrating just on prevention is too limited. Examining positive behaviours also teaches us how to encourage good leader support behaviours, which may lead to happier, more productive workers and organisations in the future (e.g., Hämming, 2017; Kossek et al., 2011; Mor Barak et al., 2009). Since general supervisor support, FSSB, and sleep leadership are all positive leader support behaviours, this research will be the first to investigate how leader sleep influences these activities. A distinctive aspect of this research is the inclusion of a constellation of support behaviours as outcomes. The little study on this subject usually looks at one kind of leader behaviour as a result of leader sleep. For instance, Barnes and colleagues (2020) looked at immoral leadership as the only result at the leader-level. Other examples of independent effects include aggressive leadership style or abusive supervision (Barnes et al., 2015; Tariq et al., 2019). (Guarana & Barnes, 2017). A research by Olsen and colleagues (2016) that looks at the relationship between leader sleep and future transformational and transactional leadership styles serves as an example of an exception. The current study will further knowledge by investigating the effects of leader sleep on a number of distinct, positive leader support dimensions (general supervisor support, FSSB, and sleep leadership).
The complexity of leader conduct in the workplace is oversimplified if just one form of leader behaviour is examined as a consequence. Theoretically, it is crucial to evaluate a constellation of leadership behaviours since doing so will help us create a more accurate representation of the numerous facets of effective leadership behaviour. This constellation would also reveal to researchers whether leaders are more likely to give up one form of support over the other when they are experiencing sleep restriction, and whether nonwork domain or work domain support may be more difficult for leaders to provide when experiencing sleep restriction. Practically speaking, investigating a constellation of positive leader behaviours informs next initiatives meant to encourage a particular positive leader behaviour. For instance, companies can think about creating policies and processes that safeguard and promote healthy sleep among the leader population if FSSB is valued and proven to be especially vulnerable to sleep loss.
Second, since sleep is important for restoring the resources used throughout the day, previous research points to resource-based processes as the connection between sleep and future leadership behaviour. According to a large body of studies, ego depletion—or the variation in our capacity to withstand desires and impulses as a result of a shortage of resources—is the underlying mechanism (e.g., Barnes et al., 2011; Barnes et al., 2015). Ego depletion theory's central tenet is that when resources are used up, a condition of depletion develops that renders a person unable or unwilling to exercise control over their conduct (Baumeister & Vohs, 2007). However, this body of research has come under heavy fire stating that since ego depletion is too vague and fails to pinpoint the precise resources at work, using it as a theoretical framework leads to the replicability dilemma (Hagger et al., 2010; Inzlicht & Friese, 2019; Lurquin et al., 2016; Lurquin & Miyake, 2017). Furthermore, since everything might be seen as a resource, the ambiguity of ego depletion results in possibly unfalsifiable outcomes (Halbesleben et al., 2014; Lurquin & Miyake, 2017). This argument is similar to those levelled against other resource-based theories like the conservation of resources (COR) theory (Hobfoll, 1989), whose main tenet is that individuals always want to conserve and acquire "resources," putting pressure on their relationships when those things are lost. Previous research have consistently argued for sleep-driven interactions in the workplace using resource-based theories like ego-depletion and COR (e.g., Barber et al., 2012; Crain et al., 2014; Sianoja et al., 2020). Researchers have recommended for more precision and accuracy when using resource theories like ego depletion after this extensive examination of resource-based theories (Friese et al., 2019). By mentioning the resource at work in these connections, the research on sleep and organisations can advance, and practitioners may learn more about how to effectively promote sleep in the workplace (Halbesleben et al., 2014; Lurquin & Miyake, 2017). This research will use a newer, more in-depth theoretical model to explicitly evaluate the resources at work in the expected interactions in order to overcome such shortcomings of prior works (Crain et al., 2018). Therefore, by speculating that emotional weariness is a mediating mechanism in the connection between leader sleep and future leader behaviour outcomes, our research adds to the body of knowledge.
The investigation of the relationship between leader sleep quantity and quality as predictors of emotional tiredness and downstream leader support behaviours is the third contribution of the current research. The evidence suggests that sleep quantity and quality should be evaluated as separate entities since the associations between them are often weak and non-significant (e.g., Barnes, 2012; Brossoit et al., 2019; Crain et al., 2018; Litwiller et al., 2017). As a result, recent research has shown that amount and quality of sleep may interact (Barber et al., 2010; Barnes et al., 2015). Because it may show how sleep quantity and quality interact to influence downstream leader behaviours, it is crucial to look into this possibility. This will help researchers and practitioners understand how to take both factors into account when developing workplace solutions. A new strategy for public health campaigns might be developed by accumulating empirical data on the relationship between sleep quantity and quality, since the majority of existing programmes stress having at least 7 hours of sleep every night.
Over the length of the night to ensure optimum functioning. Examples include sleepeducation.org, which offers a bedtime calculator and advice for scheduling time to sleep as part of the National Healthy Sleep Awareness Project's "7 and up" campaign relating to sleep length (American Academy of Sleep Medicine, 2021). Additionally, the "HealthyPeople2030" programme from the U.S. Department of Health & Human Services was started with a near-exclusive emphasis on increasing sleep quality (United States Department of Health and Human Services, 2020). Another example is the National Sleep Foundation's annual Sleep Awareness Week, which takes place the week before Daylight Savings Time, when individuals lose one hour of the day and emphasises length (National Sleep Foundation, 2021).
However, these programmes may put greater focus on providing health information about sleep quality if enough empirical research indicate it may be more important than we previously assumed. Furthermore, the interaction between sleep duration and quality is still developing, and the results examined thus far have been ego exhaustion or psychological distress (Barber et al., 2010; Barnes et al., 2015). Therefore, it is vital to investigate how this interaction affects other results. This research will add to the body of knowledge on the interaction between sleep quantity and quality by studying emotional tiredness as a possible result. This is in response to recommendations for more investigation of this interaction effect (Barber et al., 2010; Crain et al., 2018).
Finally, by integrating employee and supervisor self-ratings of sleep leadership and FSSB results, this research contributes methodologically to the organisational literature. using the direct reports' evaluations as well as the leader's own employees will prevent overstated correlations often discovered in same source data by using several sources of information about the same components (Podsakoff et al., 2003; Podsakoff et al., 2012). Similar to that, ratings could be less prone to supervisors' self-report bias. Additionally, common technique bias is less likely to influence the significance of the findings given the longitudinal design of the trial, which included assessments at baseline (Time 1), four months after the intervention (Time 2), and nine months after baseline (Time 3). (Podsakoff et al., 2003; Podsakoff et al., 2012). Additionally, this analytical approach could provide theoretical hints to the most recent leadership and occupational health literatures. How, for instance, do the views of support by workers and leaders differ? When a leader gets enough rest and doesn't feel as worn out emotionally, do workers regard them as being more supportive of their job (general supervisor support), nonwork (FSSB), and sleep (sleep leadership)? As a result, there are methodological as well as theoretical benefits to using various sources of data, particularly leader and employee assessments of two outcomes (i.e., FSSB, sleep leadership).
I offer the work, nonwork, and sleep (WNS) model as the theoretical framework for comprehending the relationship between leader sleep and subsequent workplace actions in the parts that follow (Crain et al., 2018). In addition, I draw attention to the taxonomy of human energy published by Quinn and colleagues in 2012 and suggest emotional weariness as a mediating mechanism. I discuss the pertinent lines of research to back up the precise suggested hypotheses about the relationships between leader sleep, emotional weariness, and positive support behaviours in the sections that follow. For the conceptual model, see Figure
1. Theoretical Rationale
I use the theoretical framework developed by Crain and colleagues (2018) to help explain the proposed connection between leader sleep and subsequent support behaviours. This framework identifies the underlying mechanisms connecting the three domains of employees' lives—work, nonwork, and sleep—and describes how these domains interact (WNS). According to the WNS model, sleep has a significant impact on our attitudes, actions, and feelings in both the work and nonwork domains. I depart from previous definitions of sleep, which typically focus on deficiencies and move toward a more positive definition of sleep to highlight its role in health and well-being by using Buysse's (2014) definition of sleep health. The current study focuses on the role of sleep in emotional health and downstream positive behaviours in the work domain. In the past, organisational research on sleep has mostly focused on the key variables of sleep duration and quality (Barnes, 2012). Contrarily, Buysse (2014) contends that there are other elements that go into "excellent" sleep, making the concept of sleep health more complex and multidimensional.
In particular, the key components of sleep health are sleep duration (the total amount of sleep received during each 24-hour period), sleep satisfaction (the subjective assessment of whether or not one had "good" or "bad" sleep), sleep efficiency (the ease with which one can fall asleep and stay asleep), and sleep timing (the timing of sleep during a 24-hour period) (Buysse, 2014). For the purposes of this research, I focus on sleep length as the primary variable indicating the amount of sleep, and sleep contentment and insomnia symptoms (which are both equated to sleep efficiency) as variables of the more general concept of sleep quality. Beyond looking at a mediating influence, this research also further advice drawn from the WNS theoretical framework. According to Crain and colleagues (2018), there may be an interaction impact between the amount and quality of sleep. Until now, few research have looked at such a connection (Barber et al., 2010; Barnes et al., 2015). Consequently, academics have suggested more investigation into this impact (Crain et al., 2018). This research will look at the relationship between sleep quantity and quality on subsequent emotional weariness and subsequent work behaviours in order to follow this advice and progress this topic.
The WNS model's central thesis is that energy resources from sleep have an impact on work behaviour. The WNS model builds on Quinn and colleagues' (2012) taxonomy of human energy to propose that sleep influences behaviours, attitudes, and states via fluctuations in two specific types of energy resources: physical energy, and energetic activation. This addresses criticisms of prior theories, which have conceptualised resources broadly and without much specification (e.g., ego depletion or COR; Ganster & Rosen, 2013). Physical energy, according to Quinn and colleagues (2012), is "the ability to work." To put it another way, it is the physiological energy required for acting, moving, and thinking (Quinn et al., 2012). As opposed to this, energetic activation refers to a person's perception or sense of being energised, full of vigour, energy, or zest, which may be seen in the emotive results that follow (Quinn et al., 2012). I explicitly concentrate on energetic activation as one sort of energy resource that is directly altered by sleep in the present investigation. According to Crain and colleagues (2018), sleep quantity and quality are resources that may yield other resources, most notably energetic activation, and that they have a positive association with it.
According to studies, there is a direct correlation between energetic activity and emotional tiredness, which is a key factor in burnout (Quinn et al., 2012l; Wright & Cropanzano, 1998). Emotional fatigue is defined as a persistent condition of emotional and physical tiredness and emotions of being overextended in seminal work on this concept (Cropanzano et al., 2003; Masclach et al., 2001; Maslach & Jackson, 1984). Given the same focus on affect and individual evaluation, this concept implies that energy activation and emotional weariness are closely related.
Particularly, low energy activation is a sign or symptom of emotional weariness. Since energetic activation is the resource that might result in sensations of emotional tiredness as a condition, these notions are connected but separate from one another. Thus, changes in sleep amount and quality should lead to a corresponding increase or decrease in energetic activation, which may be noticeable as emotional tiredness.
Additionally, Crain and colleagues (2018) assert that energetic activation effects behaviour in the workplace in a favourable way. The WNS model suggests that energetic activation may serve as a connecting mechanism through which sleep impacts work domain behaviours, which is pertinent to our investigation (Crain et al., 2018). For instance, prior research suggests that the production of positive emotional resources, such as energetic activation, may impact workplace behaviours including performance, helpful behaviours, and engagement (Brief & Weiss, 2002; Crain et al., 2018). Determining how emotional tiredness may be related to leaders' subsequent workplace actions, such as general supervisor support, FSSB, and sleep leadership, is the purpose of the current research.
The links between work, nonwork, and sleep that are predicted by the WNS model, according to Crain and colleagues (2018), are expected to develop over time. Because of how sleep works, it may have both short term and long-term consequences (Litwiller et al., 2017). Sleep has a critical role in long-term brain changes (i.e., plasticity), according to research in the cognitive neuroscience field (Carskadon & Dement, 2011). These changes have been found to affect behaviour in the future (e.g., Kolb, 1995; Kolb & Gibb, 2014; Kolb et al., 2003). The research on sleep also shows that its benefits deteriorate with time as a result of sleep deprivation (Barnes, 2012). Therefore, research indicates that evaluating sleep-related associations across time is more likely to represent significant brain changes and the ensuing behavioural changes (Kolb & Gibb, 2014). In light of this evidence, Crain and colleagues (2018) have urged longitudinal strategies for organisational sleep research, such as intervals spanning months as opposed to days. Additionally, Litwiller et al. (2017) made explicit requests for additional longitudinal sleep research to advance our knowledge of the dynamic long-term mechanisms underlying the relationship between sleep and downstream consequences. In light of these suggestions, this research investigates the proposed correlations through time. In particular, I anticipate that leaders who get enough sleep at Time 1 will feel less emotionally spent at Time 2, which will lead to an increase in supportive actions for other leaders at Time 3.
The Relationship Between Sleep Quantity and Support Behaviors
Establishing a relationship between sleep duration and downstream support behaviours is the primary goal of this research. According to the WNS theoretical paradigm, sleep energy-based resources have an impact on future work domain behaviour (Crain et al., 2018). Sleep, according to Crain and colleagues (2018), is a major factor in human energy swings. This energy is required for a leader to be able to participate in downstream positive support behaviours in the workplace. Prior research has started to demonstrate a relationship between sleep duration and results for leader performance. For instance, according to Gauntley (2014), peers tend to give leaders who have discrepancies between their weekend and weekday sleep duration worse performance evaluations. When they are sleep deprived, leaders are also more prone to participate in passive avoidant leadership, especially lassiez-faire (leaders who are often absent when required; Bass & Riggio, 2010). (Olsen et al., 2016). Additionally, leaders who lack sleep are less likely to exhibit transformative leadership (Olsen et al., 2016). Such results support the central tenet of the WNS theoretical framework, which holds that the amount of sleep might influence leaders' subsequent work behaviours (Crain et al., 2018).
It's crucial to look at leader support even if research has shown associations between leader sleep and future leadership results. Between leadership and support, there are significant distinctions. Most often, leadership has been thought of as a process, while support has been thought of as a particular behavior (Antonakis & Day, 2018; Hammer et al., 2009). According to Hogan and Kaiser (2005), leadership is defined as the way a person guides a team, group, or organisation to achieve a certain common purpose. Support, on the other hand, is more likely to happen on an individual basis between the leader-employee dyad. Furthermore, bad leadership (such as harsh supervision) or favorable (e.g., charismatic leadership), but generally speaking, showing support is a good and constructive action. Regardless of leadership style, support from leaders is generally a core requirement, as shown by empirical study. For instance, it has been demonstrated that leadership support results in a variety of positive outcomes, including increased employee creativity (e.g., Cheung & Wong, 2011), decreased work-to-family conflict and family-to-work conflict, better employee sleep (e.g., Sianoja et al., 2020), higher job satisfaction (e.g., Odle-Dusseau et al., 2012), and work engagement (e.g., May et al (e.g., Nohe & Sonntag, 2014) Consequently, support is an important and beneficial aspect of leadership that is worth looking at in isolation in order to determine how to preserve and encourage such behaviours at work.
General Supervisor Support
The influence of sleep duration on three support behaviors—general supervisor support, family-supportive supervisor behaviours (FSSB), and sleep leadership—is examined in this research. In order to promote a worker's productivity at work, a leader must provide general supervisor support. General supervisor support specifically refers to actions like offering practical aid and services (i.e., instrumental support) and exhibiting compassion, encouragement, care, and trust (i.e., emotional support) for their staff members at work (House, 1981; Langford et al., 1997; Mathieu et al., 2019; Yoon & Thye, 2000). As it has been shown to reduce employee anxiety, depression, somatic complaints, emotional tiredness, and turnover, among many other things, previous research has underlined the significance of boosting general supervisor support in the workplace.
O'Driscoll et al., 2003; Siebert, 2006; Haas et al., 2020; Mor Barak et al., 2009; Since providing practical and emotional support to a team of workers demands energy and effort, sleep is likely to have an influence on a leader's tendency to engage in general supervisor support. For instance, a manager who gets more sleep is more likely to support and care for their staff members and to have the means to help them. Therefore, it is expected that sleep duration will correlate with overall supervisor support for employees (See Figure 1).
Hypothesis 1: A leader's amount of sleep at Time 1 will be positively correlated with staff members' perceptions of overall supervisor support at Time 3.
Family-Supportive Supervisor Behaviors
The FSSB architecture lays emphasis on leaders supporting their employees' non-work needs as opposed to generic supervisor assistance. FSSB is envisioned as domain-specific leadership practises that help employees succeed in both their professional and personal life (Crain & Stevens, 2018; Hammer et al., 2009). For instance, FSSB could include actions like showing concern for employees' non-work lives, providing resources to help with demands, exemplifying how to successfully balance one's own work and non-work responsibilities, or making proactive efforts to improve employees' capacity to balance their work and non-work demands (Hammer et al., 2011). A recent review found that FSSB can have a variety of positive effects on work outcomes like job satisfaction, performance, and the quality of leader-follower relationships as well as on health outcomes like sleep (Berkman et al., 2010), employee stress (e.g., Hammer et al., 2013), and burnout (e.g., Koch & Binnewies, 2015) (Crain & Binnewies).
(2018) Stevens A leader who supports families is one who "understands the need of the employee to achieve balance between work and nonwork duties" (Thomas & Ganster, 1995, p.7). A leader's capacity to participate in FSSB is likely to be impacted by their sleep since such behaviours may be difficult because they need emotion, empathy, and proactivity to be successful. Sleep deprivation has been linked to impairments in emotional control, empathy, and proactivity, according to studies by Palmer & Alfano (2017), Guadagni et al. (2014), and Guadagni et al. (2017). (e.g., Schmitt et al., 2017). For instance, a boss who gets adequate sleep may be more empathetic and proactive in addressing and resolving a worker's non-work need. Therefore, knowing how sleep contributes to FSSB can help you promote these positive habits in the workplace (See Figure 1).
Hypothesis 2: Leader and employee reports of FSSB at Time 3 will be positively correlated with leader sleep quantity at Time 1 and vice versa.
Although FSSB refers to supervisor actions that help workers balance professional and personal obligations, this concept does not directly address the area of sleep. Crain and colleagues (2018) claim that despite the fact that sleep takes up a large amount of a 24-hour period, previous study analysing the domains of a working person's life often neglected sleep as an important area. Sleep leadership, as opposed to FSSB, refers to supporting actions that specifically address the sleep domain of an employee's life (Gunia et al., 2015). Employers are assisted in achieving their sleep objectives by leaders who practise sleep leadership activities concern for the sleep wellness of employees. In general, sleep leadership might play a supporting role in enhancing worker sleep (Gunia et al., 2015). It is crucial to encourage sleep leadership at work, since this behaviour has been associated with better employee sleep, supervisor sleep knowledge, organisational environment, and a reduction in employee melancholy symptoms (Adler et al., 2021; Gunia et al., 2015). According to a recent study on short sleep duration among working people in the United States, managers and companies who encourage employees to get enough sleep may see advantages including enhanced workplace productivity, lower healthcare expenses, and better workplace safety and health (Khubchandani & Price, 2020). When a leader's sleep is suffering, they may not know how to have good sleep in their own lives and may be less likely to be able to teach their staff that knowledge. This may affect their capacity to give sleep leadership to their colleagues. Additionally, similar to FSSB, demonstrating care for employees' sleep is probably difficult, particularly when a leader's own sleep is compromised. Indeed, sleep research has shown that sleep limitation affects how much effort is allocated (Massar et al., 2019), indicating that leaders may be less inclined to devote effort to performance objectives like promoting and preserving staff sleep. For instance, executives who do not get enough sleep may find it difficult to support their staff members with sleep since their own sleep is being compromised. However, unlike FSSB, leader sleep may be linked to future sleep leadership actions because if a leader does not exhibit concern and care for their own sleep, it is extremely unlikely that they would do the same for others their staff members' sleep. Therefore, it's crucial to comprehend how leader sleep may affect their capacity to promote and care for employee sleep (See Figure 1).
Hypothesis 3: Leader and employee reports of sleep leadership at Time 3 will be positively correlated with leader and employee sleep quantity at Time 1.
Emotional Exhaustion as a Mediator
The second goal of this research is to determine how and why leadership behaviour and sleep quality could be connected. The WNS theoretical model's fundamental concept is that sleep affects behaviour at work by using energy sources like energetic activation (Crain et al., 2018). According to Quinn et al. (2012), emotional weariness is a symptom of low energetic activity and happens when resources are depleted or inadequate to satisfy expectations (Hobfoll, 1989). The physiological process of sleep, which is crucial, may help with the replenishment of such resources (Barnes, 2012; Toker & Melamed, 2017). Accordingly, theory predicts that the amount of sleep has a negative connection with energetic activation, leading to higher emotional tiredness (Crain et al., 2018).
Additionally, it is conceivable that emotional weariness is the cause of how leader sleep affects later pro-social activities. Affective constructs like emotions, moods, or dispositions have been connected to energetic activation and emotional weariness (Crain et al., 2018; Quinn et al., 2012). Particularly, energetic activation and emotional fatigue are linked to future emotions, moods, or dispositions; specifically, one is more likely to struggle with emotion management, leading to an increase in negative affect, when energetic activation and emotional tiredness are low (Lam et al., 2010; Quinn et al., 2012; Wright & Cropanzano, 1998). Similarly, People who are emotionally spent are prone to feel emotionally stretched out (Maslach & Jackson, 1981). Sleep and job outcomes may be linked via an affective-based construct, according to organisational health academics (Barnes, 2012; Henderson & Horan, 2021; Mullins et al., 2014). Positive leader support behaviours, such as general supervisor support, FSSB, and sleep leadership, depend on the leader believing they have the energy to show concern for the employee's work and personal life, both of which are inherently emotional and are expected to be affected by their emotional exhaustion (Quinn et al., 2012). As a result, the current model emphasises emotional weariness as a mediator in order to improve our comprehension of the fundamental affective mechanism linking sleep to leader support behaviors based on experimental sleep research, there is strong evidence connecting sleep to affective factors like mood (i.e., affective states without a clear cause and low intensity; Daus et al., 2020), emotion (i.e., affective reactions with a clear cause, short duration, and high intensity; Frijda, 1993), and interpersonal functioning (i.e., empathy toward others and quality of relationships; Killgore et al., 2008). People exhibit considerable mood abnormalities, mental tiredness, and emotional complaints (such as sadness, anxiety, and vigour), for instance, when their sleep is limited in an experimental context (e.g., Banks & Dinges, 2007; Dinges et al., 1997; Friedmann et al., 1977; Goel et al., 2009; Johnson & MacLeod, 1973; Short & Banks, 2014). Laboratory studies have also shown that sleep deprivation can affect one's capacity to recognise and appropriately respond to another person's emotions (e.g., Amicucci et al., 2021; Tempesta et al., 2020; Van der Helm et al., 2010), which is important because relationship building is a requirement for effective leadership (Riggio & Reichard, 2008). Additionally, it has been shown that experimental sleep restriction reduces one's propensity to respond in a way that promotes successful social interaction and weakens one's capacity to stifle unpleasant feelings (Kahn-Greene et al., 2006).
Finally, Killgore and colleagues (2008) discovered that experimentally controlled sleep restriction decreased behavioural coping, interpersonal functioning, stress management abilities, and intrapersonal functioning (such as assertiveness, independence, and relationship quality) (e.g., action orientation). Results from non-experimental research support the notion that sleep may have an influence on emotional tiredness. Reduced sleep duration, for instance, has been linked to interpersonal ineffectiveness and emotional weariness (Ekstedt et al., 2006; Jansson-Fröjmark & Lindblom, 2010; Litwiller et al., 2017; Nowack, 2017; Rosen et al., 2006; Söderström et al., 2012). Therefore, lack of sleep might harm interpersonal or emotional abilities that are essential for a leader to exhibit supportive actions.
The relationship between emotional tiredness and leadership conduct has also been shown in earlier research. For instance, it has been shown that emotional tiredness is a significant risk factor that might lead to abusive leader conduct (e.g., Fan et al., 2020; Lam et al., 2017). The association between sleep deprivation and magnified negative emotional reactivity, such as heightened irritability and affective volatility, is supported by studies on sleep physiology (Horne, 1985; Dinges et al., 1997; Walker et al., 2009; Zohar et al., 2005). Emotional weariness serves as a connecting variable between work instability and transformative leadership behaviour, according to an empirical research by Qian and colleagues (2020). This body of evidence backs up the hypothesis that emotional tiredness may have an impact on subsequent behaviour in the workplace (Crain et al., 2018). Little study has been done on emotional tiredness of the leader, and the majority of the literature on emotional exhaustion and leadership focuses on how leader styles and actions might affect employee emotional weariness. Studies show that emotional tiredness in leaders has serious repercussions for both workers and the business, including employee well-being and job performance (e.g., Lam et al., 2010).
The current research will investigate the function of emotional tiredness as a connecting mechanism between sleep quality and support outcomes, building on this theoretical and empirical base. Given that these activities have an affective meaning, I particularly feel that emotional weariness is directly linked to leader support behaviours including general supervisor support, FSSB, and sleep leadership (See Figure 1). To assist an employee's needs, for instance, leaders must sympathise with their worries and use their own resources (e.g., Hammer et al., 2011). As a result, emotional tiredness may have a particularly negative effect on a leader's capacity to help their workers.
Hypothesis 4a: The connection between leader sleep quantity at Time 1 and employee ratings of overall supervisor support at Time 3 will be positively mediated by leader emotional weariness at Time 2.
Hypothesis 4b-c: The association between leader sleep quantity at Time 1 and leader and employee reports of FSSB at Time 3 is positively mediated by leader emotional tiredness at Time 2 and by leader and employee reports of FSSB at Time 3.
Hypothesis 4d-e: A leader's emotional tiredness at time two will act as a mediator between the leader's amount of sleep at time one and the leader and employee reports of sleep leadership at time three.
Interaction between Sleep Quantity and Sleep Quality
This research also seeks to investigate how this link could alter if sleep quality is taken into account in the model, building on assumptions that suggest that leader support actions and sleep quantity are associated through emotional weariness. According to Barnes (2012), previous research has often looked at sleep quantity and quality as complementary aspects of sleep, which means that each sleep dimension's impacts must be considered separately and cannot be replaced. The WNS model, however, references other studies that show a link between sleep quantity and quality that interacts (Crain et al., 2018), indicating that there may possibly be a multiplicative impact between the constructs. Barber and colleagues (2010) discovered that, specifically, sleep quantity and quality interact to buffer against psychological strain as a result, such that the association between sleep quantity and psychological strain is decreased under circumstances of good sleep quality. In addition, although Barnes and colleagues (2015) predicted that sleep quantity and sleep quantity would have additive effects on ego depletion and subsequent leader behaviour, the results showed that the association between sleep quantity and daily ego depletion was weaker under conditions of high sleep quality, providing empirical support for the predicted interaction. Researchers have urged more investigation of this influence on other health and wellness outcomes due to the uniqueness of this association. Future study, according to the authors of the WNS framework, should focus on dissecting and comprehending the particular resources and processes that are impacted by various constructions of sleep (Crain et al., 2018). In order to further the body of research on the relationship between sleep amount and quality, this study looks at emotional weariness as a possible result. For instance, leaders who get enough sleep and feel refreshed and content with it will feel less emotional weariness than those who don't get enough sleep or have terrible sleep. Overall, I anticipate that getting more sleep reduces emotional weariness, and this link is strengthened when getting better sleep (H5; See Figure 1).
Hypothesis 5: The fifth hypothesis states that leader sleep quality at Time 1 will moderate the relationship between leader sleep quantity at Time 1 and leader emotional exhaustion at Time 2, such that the negative relationship between sleep quantity and emotional exhaustion will be strengthened under high (versus low) sleep quality circumstances.
Moderated Mediation Effects
The current study hypothesizes that this relationship takes place through a moderated mediating framework such that the indirect effect of sleep quantity on three distinct positive leader support behaviours via emotional exhaustion is enhanced by sleep quality. Overall, empirical and theoretical evidence suggests that sleep can impact work-related attitudes, behaviours, and states (Crain et al., 2018). A leader's tendency for making decisions is particularly influenced by high-quality and enough amounts of sleep, It is substantially less likely that participants would experience downstream emotional weariness and a subsequent decline in supportive actions at work. For instance, a leader who regularly gets adequate sleep and feels refreshed won't feel as emotionally stretched, making them better able to help staff members at work. Therefore, it is anticipated that emotional weariness, which results from insufficient sleep, moderates the indirect impact of insufficient sleep on general supervisor support, FSSB, and sleep leadership (H6; See Figure 1).
Hypotheses 6a-b: Leader emotional weariness at Time 1 will at Time 2 lessen the indirect influence of leader sleep quantity at Time 2 on employee assessments of overall supervisor support.
Hypothesis 7a-d: The indirect impact of leader sleep quantity at Time 1 on leader and employee assessments of FSSB at Time 3 will be moderated by leader sleep quality at Time 1 via leader emotional weariness at Time 2.
Hypothesis 8a–d: The indirect impact of leader sleep quantity at Time 1 on leader and employee assessments of sleep leadership at Time 3 will be moderated by leader sleep quality at Time 1 via leader emotional weariness at Time 2.
Procedure and Participants
The information was gathered as a part of a wider sleep and health intervention research that spanned from 2017 to 2020. I focused on a sample of CEOs paired with their corresponding direct workers. Participants in this research were mostly full-time workers from one state in the Pacific Northwest of the United States who were members of the Army and Air National Guard. Participants held a broad range of jobs, but the majority were leaders and workers in the maintenance, logistics, finance/supply, and human resources departments. Three time periods were used to complete surveys: baseline, four months after the intervention, and nine months after the intervention (Time 3). I thus looked at emotional tiredness at Time 1, sleep length and quality at Time 2, and support behaviours at Time 3 (general supervisor support, FSSB, and sleep leadership). I controlled for the intervention indicator in all analyses since the intervention was not of significant interest in this research, as I go into more detail about below.
For comprehensive details on recruiting and study logistics, please check Hammer et al. (2021). When it came to gathering data for this thesis, the research team originally collaborated with the National Guard's headquarters and received organisational charts, breakdowns of units and their commanders, and contact information for those leaders. A person-of-contact within the National Guard was found who linked the study team with the relevant commander for smaller units and the Air branch. Using this data, leaders were emailed and given a debriefing on the research. The study team distributed an online survey to unit leaders' personal email addresses using REDCap (Research Electronic Data Capture), asking them to answer.
Additionally, managers were required to provide information about the research and a link to participate in it in an email sent to full-time staff members who directly reported to them. In addition to receiving an online survey via REDCap, participants who worked at least 32 hours per week were eligible to join up. All surveys were filled out on days off from work.
In their online survey, employees were prompted to identify their immediate supervisor (i.e., the leader that they report to if they needed to take time off work).
After all the data was gathered, participating leaders were matched with the corresponding workers based on this information and the organisational structure that was provided to the research team at the start of the study. With the help of individual unit leaders, the research team developed a final list of leaders for each unit based on who in the unit took part in the survey and was connected to workers. As a result, the final data set contains workers who are connected to each leader, where each leader may have a single direct employee or many. Specifically, as a consequence of this approach, one to thirteen workers were matched to each leader in the dataset.
Over the course of a year, participants were required to complete three 45-minute online questionnaires; the questions were the same for every participant (leaders and employees). The possible prize for finishing all survey waves was $75, with participants being given a $25 gift card as an incentive for completing each individual survey. After that, research workers went to Army and Air bases to personally inform employees about the study and sign up those who had not yet done so. In-person and online recruiting strategies were used to boost participation and reduce research attrition. Before beginning the research, all participants completed informed consent forms, and the protocol was approved by the Institutional Review Boards of the lead investigators' respective universities.
To construct the appropriate dataset for this research, data from several measuring occasions and sources (such as leaders and workers) were combined. Leaders who couldn't be paired with at least one participating employee were taken out of the dataset based on survey replies. Employees who could not be matched with a participating leader were similarly eliminated from the dataset. N = 178 and N = 393 were the final sample sizes for matched leaders and workers, respectively.
The majority of leader participants were male (80.9%), married (82.6%), Caucasian (84.3%), and in their middle age range of 40.8 years (SD = 7.30). The average number of children per leader was two (SD = 1.4), and 41.6% of them had graduated from college. Leaders worked an average of 44.83 hours per week (SD = 5.31), worked a normal day shift (89.3%), and had an average of six direct subordinates (SD = 6.18). Their average tenure was 5.39 years (SD = 5.80). The majority of employee participants were male (74%), married (65.6%), Caucasian (81.9%), and aged 35.8 on average (SD = 8.86). The average number of children per employee was two (SD = 1.4), and 43.5% of workers had merely attended some college or technical school without earning a degree. Employees worked an average of 42.37 hours per week (SD = 5.0), on a normal day shift (81.2%), and had an average tenure of 4.36 years (SD = 5.56).
Leader sleep quantity
At Time 1, leaders were asked to rate the length of their sleep in the previous month. Leaders were instructed that their responses should represent their typical sleep duration for the majority of days and nights during the previous month. The Pittsburg Sleep Quality Index's two items were used to measure the length of time spent sleeping (PSQI; Buysse et al., 1989). The questions were "When have you typically woken up in the morning over the last month?" and "When have you typically gone to bed during the past month?" (see Appendix). Leaders responded to each question by stating the time they went to sleep and woke up, including the hour (01-12), minute (00-59), and which 12-hour period of the day (AM/PM). To calculate the length of the leader's slumber, these things were employed (i.e., a difference score between when the leader reported they went to bed and when they woke up).
Leader sleep quality
At Time 1, leaders were questioned about their recent sleep hygiene and the degree to which it affected them. Eight items from the PROMIS Sleep Disturbance scale (Cella et al., 2010; PROMIS, 2016; Yu et al., 2012) were used to evaluate the sleep quality construct and were divided into two measures based on the findings of a confirmatory factor analysis. The dimension of sleep discontent is represented by four items. "I was happy with my sleep," as an example. With the exception of one question, which was graded on a 5-point scale with a different anchor (1 = extremely bad, 5 = very excellent), all of these items were assessed on a 5-point scale (1 = not at all, 5 = very much) (Cronbach's alpha =.87). The insomnia symptoms dimension is represented by the following four items. I had problems sleeping, to provide one example. On a 5-point scale, items were assessed (1 being never and 5 being always; Cronbach's alpha =.82). (See Appendix). Scale scores for both sleep quality dimensions were generated using the HealthMeasures (2021) scoring system and a t-score transformation metric, as is advised for assessments of this kind. To better examine distributions and percentiles from this sample across what we know from earlier research to be the norm (i.e., average) of sleep quality in the U.S., this t-score modification is required. Additionally, since the ratings are obtained from IRT, this choice is seen to be the most accurate (using response pattern scoring). This technique, which is advised for employing subsets of items, also deals with missing data. Overall, higher scores indicate more insomnia symptoms for one component and more sleep dissatisfaction for the other dimension.
Leader emotional exhaustion
At Time 2, leaders were asked to rate how often they had felt emotionally spent during the previous month. The scale included three items and was scored on a range of 1 to 7 (where 1 meant "never or nearly never" and 7 meant "often or virtually always"; Cronbach's alpha =.78). "I believe I am not capable of engaging emotionally in colleagues," is one such item (Shirom & Melamed, 2006). (See Appendix). If at least 75% of the questions on each scale were answered, mean imputation was used to construct scale scores for leader emotional tiredness. Otherwise, if 75% of valid item replies were missing, all scales and items were subject to list wise elimination.
Employee-rated general supervisor support
At Time 3, workers rated their agreement with each statement using three items on a 5-point scale (1 being strongly disagree, and 5 being strongly agree; Cronbach's alpha =.78). "My boss can be counted upon when things get rough in my job" is an example item (Yoon & Lim, 1999). (See Appendix). If at least 75% of the items on each scale were responded to, mean imputation was used to construct scale scores for employees rating general supervisor support.
Leader- and employee-rated family-supportive supervisor behaviors
Leaders assessed how much they concurred that they displayed FSSB at Time 3. For the current study, the short form 4-item FSSB measure (Hammer et al., 2013) was used (Cronbach's alpha =.89). On a 5-point scale, leaders were asked to react to four questions (1 being strongly disagree, and 5 being strongly agree). 'I make my subordinates feel comfortable talking to me about their problems between work and non-work,' is an example item (See Appendix).
The same scale was used to ask employees who were related to each leader to rate how much they concurred that their immediate leader displayed FSSB at Time 3. Four questions with ratings on a 5-point scale and responses from the employees were collected (Cronbach's alpha =.95). "Your boss makes you feel comfortable talking to him/her about your difficulties between work and non-work," as an example, is one of the items on the list (See Appendix). If at least 75% of the items on each scale were answered, mean imputation was used to construct scale scores for the leader and staff ratings of FSSB.
Leader- and employee-rated sleep leadership
At Time 3, leaders rated their level of agreement with eight statements using a 5-point scale (Gunia et al., 2015); 1 meant they strongly disagreed, and 5 meant they strongly agreed (Cronbach's alpha =.88). "I advise my subordinates to get enough sleep" is an example item (See Appendix).
The extent to which each leader's direct reports concurred that they displayed sleep leadership practises at Time 3 was also inquired of them. On a 5-point scale (1 = never, 5 = always), employees answered to eight items (Cronbach's alpha =.94). "My supervisor encourages subordinates to get enough sleep" is one example item (Gunia et al., 2015). (See Appendix). If at least 75% of the items on each scale were answered correctly, mean imputation was used to construct scale scores for the leader and staff ratings of sleep leadership.
Following Bernerth and Aguinis' (2016) discussion on the usage of statistical control variables, a set of control variables for inclusion were chosen based on theory and prior research (See Appendix). This study specifically departed from the purification principle (Spector & Brannick, 2011) since current research reveals that control variables may be damaging analyses by altering the meaning of the association, lowering power, decreasing explained variance, and limiting degrees of freedom (Bernerth et al., 2018). Additionally, some researchers even assert that include too many control factors could lead to inaccurate conclusions and outcomes that are impossible to replicate, impeding the advancement of science (Becker et al., 2016). To account for alternative explanations and false links in the model, selected control variables should be theoretically and empirically related to variables of interest. The following section summarises conceptually pertinent control variables that are incorporated into analyses, in accordance with suggestions made by prior researchers (Aguinis et al., 2019; Aguinis & Vandenberg, 2014; Bernerth & Aguinis, 2016; Bernerth et al., 2018). Additionally, all analyses were conducted both with and without control variables, and standard descriptive statistics, including correlations and significance levels, were presented for all control variables (See Table 1). (Aguinis & Vandenberg, 2014; Bernerth & Aguinis, 2016; Bernerth et al., 2018). All data are provided with the inclusion of controls because there were no significant changes in the outcomes; however, I still go on to explain the theoretical justification for doing so below.
Additionally, I complied with current suggestions from leadership scholars about the inclusion of demographic control factors because the primary focus of this study was on workplace leaders. According to a systematic review by Bernerth and colleagues (2018), the relationship between leadership-relevant dimensions and frequently employed control variables like age, gender, tenure, or education has an unimpressive effect size. The inclusion of such demographic controls, according to Bernerth and colleagues (2018), is specifically suggested to be unsupported by theory and simply based on out-of-date beliefs about control variables, which has a major negative impact on analyses. Therefore, in this study, proxy demographic control variables like the leader's age, gender, tenure, or education were not included (Bernerth et al., 2018).
Work-Related Controls. According to empirical study, some work qualities may have an effect on sleep. For instance, shift work has been linked to poor sleep (e.g., Kerstedt, 2003; Van Dogen et al., 2006). This connection may be explained by the fact that our biological clock's circadian rhythm creates times of day that are better for sleeping. Additionally, it's likely that in the current study, leaders or employees who work shifts that differ from the standard day shift may interact with one another less, leading to false results because the study's hypotheses depend on contact between leaders and employees. Work schedule was therefore included as a control variable in studies since shift work may affect both leader sleep as well as leader and employee ratings of positive support behaviours.
There is a small subset of the sample that can be classified as high risk, despite the fact that the bulk of the sample members work in fields like human resources or finance/logistics. High-risk vocations are those that carry a significant and unpredictable risk (Gunia et al., 2015). Law enforcement (Russell, 2014), mining, manufacturing, agriculture (Earnest et al., 2018), and military service are a few examples of these professions (Elliman et al., 2020; Gunia et al., 2015). Due to heightened stress, those who work in high-risk occupations have been demonstrated to be especially vulnerable to sleep deficits (Akerstedt & Wright, 2009; Gunia et al., 2015; Linton et al., 2015; Seelig et al., 2016; Seelig et al., 2010). Thus, the impact of sleep on various tasks may vary. Additionally, it's feasible that various support behaviours will be more or less crucial for various vocations. For instance, in the small high-risk segment of the sample, leaders may need to exhibit sleep leadership support behaviours more so than leaders in the human resources division. Both the Army and the Air National Guard, which each perform different sorts of labour that may impair sleep and supportive behaviours, were represented in the participant sample. According to Whealin and colleagues (2015), the Army branch of the National Guard had worse health outcomes than the Air branch, including greater rates of post-traumatic stress disorder and more severe physical and mental health problems. Because of the wide range of jobs in this sample, the branch of service (Army, Air Force, etc.) was added as a control variable.
The larger study was an intervention in a randomized-controlled trial, to sum up. The intervention indicator (0 = normal practise, 1 = intervention) was included as a control variable because the intervention is not an important variable for the current investigation.
Family-Related Controls. According to empirical studies, people with kids at home sleep for shorter amounts of time than people without kids (e.g., Burgard & Ailshire, 2012; Hagen et al., 2013; Khubchandani & Price, 2020; Tienoven et al., 2014). The relationship between the two may be explained by the fact that time is limited and people with children must devote their limited free time to looking after them, which may interfere with their ability to sleep. Furthermore, having children may make leaders more sympathetic to the needs of their workforce outside of work, which could translate into more support outside of the office (e.g., FSSB, sleep leadership). Number of children was added as a control variable since it may affect both leader sleep and future support behaviours. Similar to this, people who are responsible for providing care for elderly people may experience inadequate sleep (Burch, 2019; Dugan et al., 2020), as they similarly report having less time to sleep (American Psychological Association, 2012; Caruso et al., 2006). Additionally, people who have eldercare responsibilities report poor sleep quality because they frequently have sleep interruptions and struggle to go to sleep at night (Hoyt et al., 2021). The similar theoretical justification can be given for childcare and eldercare obligations, arguing that leaders who have eldercare obligations may be more sympathetic toward employee nonwork demands and, as a result, exhibit more supportive behaviours at work. Eldercare obligations were therefore added as a control variable.
Data Cleaning And Preliminary Analyses
Data input errors, missing values, and outliers were checked for before the primary analysis were carried out. All essential variables' scale scores were generated using mean imputation, as indicated in the section on measures, and were subject to listwise deletion if a certain threshold wasn't reached. After that, SPSS was used to check if the multilevel multiple regression assumptions (i.e., normality, linearity, independence of errors, lack of multicollinearity, heteroscedasticity) were met for the suggested key variables (i.e., predictors, moderators, mediators, outcomes, and controls).
The data were originally checked for univariate outliers by viewing frequency distributions, box and whisker plots, histograms, and box and whisker plots. There were two outliers on the sleep duration and sleep dissatisfaction variables for the predictors and mediators, and five outliers on the insomnia symptoms measure. Three outliers on general supervisor support, four outliers on leader evaluations of FSSB, five outliers on employee ratings of FSSB, and four outliers on employee ratings of sleep leadership were present for the outcome variables. However, after careful inspection of the outlier values, it became clear that there was no theoretical justification for keeping these outliers, hence the detected univariate outliers were kept. Relatives analysis, Cook's D, Mahalanobis' distance (24.32 based on degrees of freedom), centred leverage values, and scatterplots were used to look for multivariate outliers. Evidence suggested that multivariate outliers were absent.
After that, the normality and linearity of the data were evaluated. Employee assessments of sleep leadership and emotional weariness were both slightly positively skewed, despite the fact that no adjustments improved estimations of kurtosis and skewness. The initial variables were kept because of this. The distribution of sleep duration, symptoms of insomnia, sleep discontent, and leader-ratings of sleep leadership was typical. Employee ratings and general supervisor support for FSSB were skewed adversely. Square root transformation was used to enhance the skewness and kurtosis of negatively skewed data, however the outcomes of modelling with and without the transformed variables were not noticeably different. As a result, analysis and reporting both used the original, untransformed numbers. All other variables satisfied the requirements of normality and linearity. Variables matched the heteroscedasticity assumptions and showed independent of mistakes, according to both histograms and scatterplots (i.e., of associations as well as residuals). Finally, psychometric tests were performed to determine the measures' reliability and validity. For each measure, Cronbach's alpha was calculated.
Multilevel Modeling. In order to assess the level of reliance inside work groups, intraclass correlation coefficients (ICCs) were computed using the organisational group variable due to the layered structure of the data, in which participating employees worked within work groups under the direction of leaders. Using multilevel modelling in later analyses can be determined by computing ICCs. A significant amount of work group dependence in the outcomes is shown by the computed ICCs for emotional exhaustion (ICC =.09), leader-ratings of FSSB (ICC =.27), and leader-ratings of sleep leadership (ICC =.16). Due to convergence problems, it was not possible to compute ICCs for general supervisor support, employee evaluations of FSSB, or sleep leadership. This is probably because there was insufficient dependency among the task groups. Multilevel modelling was performed because it is a more cautious approach to nested data and because important outcome leader variables had relatively high ICCs.
Descriptive Statistics. All research variables, including the control variables, are included in Table 1 along with descriptive statistics and bivariate correlations. At Time 1 (SD = 0.94), leaders averaged 7.36 hours per day of sleep over the previous month, and at Time 2 (M = 2.07, SD = 1.00), leaders averaged a comparatively low level of emotional exhaustion. At Time 3, the majority of workers (M = 4.23, SD = 0.84) firmly believed that their managers gave general supervisor assistance. In addition, leaders and employees differed on whether the leader offered sleep leadership at Time 3 (leader, M = 2.61, SD = 0.78; employee, M = 2.51, SD = 1.0); nevertheless, both groups concurred that the leaders supplied FSSB at Time 3 (leader, M = 4.10, SD = 0.49; employee, M = 4.11, SD = 0.09).
In order to understand the fundamental structure of interactions between variables, bivariate correlations were also examined. Leader sleep duration at Time 1 or leader sleep pleasure at Time 1 did not significantly correlate with either leader emotional weariness at Time 2 or leader sleep duration. Leader emotional weariness at Time 2 was substantially and favourably correlated with leader sleeplessness symptoms at Time 1 (r = 0.16, p .01). Interestingly, leader insomnia symptoms at Time 1 were not linked with any of the employee-rated support outcomes at Time 3 but were significantly and negatively associated with leader assessments of FSSB (r = -0.12, p .05) and sleep leadership (r = -0.14, p .01). Additionally, leader emotional weariness at Time 2 was not associated with leader ratings of sleep leadership at Time 3 (r = -0.32, p .01), but it was strongly and adversely correlated with leader ratings of FSSB at Time 3 (r = -0.32, p .01). At Time 3, however, employee perceptions of general supervisor support (r = -0.16, p .05), FSSB (r = -0.13, p .05), and sleep leadership (r = -0.14, p .05) were substantially and adversely connected with leader emotional tiredness at Time 2. Interestingly, there was no discernible correlation between the leader and employee assessments of FSSB at Time 3. Additionally, there was no discernible correlation between leader and employee assessments of sleep leadership at Time 3. Unsurprisingly, there was a high correlation between all employee ratings of general supervisor support, FSSB, and sleep leadership at Time 3 (p .01).
The association between Time 1 leader sleep duration and Time 3 leader behaviour outcomes (i.e., leader-reports of general supervisor support, leader and employee-reports of FSSB, and leader and employee-reports of sleep leadership) as mediated by Time 2 leader emotional exhaustion was explored in the main analyses through a series of multilevel moderated-mediation models. Time 1 leader sleep quality (i.e., sleep dissatisfaction, insomnia symptoms) was also evaluated. Finally, I looked at how sleep quality affected the leader outcomes that were mediated by leader emotional tiredness and leader sleep length. The branch of service (Army vs. Air), condition (normal practise vs. intervention), number of children, eldercare, and job schedule were all included as controls in the models (i.e., daytime vs. other). Multilevel fully-saturated path analyses were performed throughout analysis using Mplus Version 8. (Muthen & Muthen, 2018). I ran a series of five moderated mediation models due to the complexity of the overall model with all variables included and the resulting convergence issues. In these models, the predictor (i.e., sleep duration), both moderators (i.e., sleep dissatisfaction and insomnia symptoms), the mediator (i.e., emotional exhaustion), and only one outcome (i.e., general supervisor support, leader-ratings of FSSB, employee-ratings of FSSB.
According to Hypothesis 1, the amount of sleep a Time 1 leader gets at Time 3 would be positively correlated with the employees' perceptions of their overall supervisor support at Time 1. There was no significant correlation between leader sleep duration at Time 1 and employee ratings of general supervisor support at Time 3 after adjusting for all other factors in the model (b = 0.62, SE = 0.05, p = 0.22, 95% CI [-.048,.146]). A positive correlation between the leader and employee reports of FSSB at Time 3 and the leader's sleep length at Time 1 was predicted by Hypothesis 2. After accounting for all other factors in the model, it was shown that there was no significant correlation between leader sleep duration at Time 1 and leader or employee reports of FSSB at Time 3 (b = -0.01, SE = 0.08, p = 0.95, 95% CI [-.13,.20]). According to Hypothesis 3, Time 1 leader sleep duration would be positively correlated with Time 3 leader and employee reports of sleep leadership. After adjusting for all other factors in the model, there was no correlation between the length of the leader's sleep at Time 1 and the leader's or the employee's reports of sleep leadership at Time 3 (b = -0.05, SE = 0.11, p = 0.64, 95% CI [-.27,.17]). As a result, None of Hypotheses 1-3 were supported. For a list of the direct impacts, see Table 2.
According to hypotheses 4a through 4e, the positive relationship between a leader's sleep duration at Time 1 and employee reports of general supervisor support at Time 3 as well as between a leader's sleep duration at Time 1 and employee reports of FSSB and sleep leadership at Time 3 would be moderated by the leader's emotional exhaustion at Time 2. The importance of indirect effects was assessed via bootstrapping with 5,000 bootstrapped samples. This method has the advantage of producing standard errors that are robust to outlier values and distributional problems (Chernick et al., 2014). Statistical significance was assessed using 95% confidence intervals that excluded zero. The provided results are typical after accounting for all other model parameters.
Results showed that leader emotional weariness at Time 2 had a non-significant indirect effect on employee ratings of overall supervisor support at Time 3 (indirect effect = 0.01, 95% CI [-.02,.06]). As a result of leader emotional weariness at Time 2, there was a non-significant indirect effect of leader sleep duration at Time 1 on leader-rated FSSB at Time 3 (indirect effect = 0.00, 95% CI [-.01,.05]). Through leader emotional tiredness at Time 2, there was a non-significant indirect effect of leader sleep duration at Time 1 on employee-rated FSSB at Time 3 (indirect effect = 0.01, 95% CI [-.01,.05]). Additionally, there was a negligible indirect effect of leader emotional tiredness at Time 2 on leader-rated sleep leadership at Time 3 (indirect effect = 0.00, 95% CI [-.01,.03]).
Last but not least, leader emotional exhaustion at Time 2 had a non-significant indirect effect on employee ratings of the leader's sleep leadership at Time 3 (indirect effect = 0.02, 95% CI [-.02,.06]). As a result, none of Hypotheses 4a–e were verified. For a summary of indirect impacts, see Table 3.
I suggested a potential interaction between leader sleep duration and leader sleep quality (i.e., sleep discontent and insomnia symptoms), therefore interaction terms were made with values of the predictor and moderator that were grand mean centred to prevent problems with multicollinearity (Tabachnik & Fidell, 2007). The relevant control variables were utilised depending on the outcome (e.g., employee or leader ratings) (i.e., employee work schedule vs. leader schedule) 1. The hypotheses 5a and 5b proposed that the leader's sleep quality at Time 1 (i.e., sleep dissatisfaction [5a] and symptoms of insomnia [5b]) would moderate the relationship between the leader's sleep duration at Time 1 and the leader's emotional exhaustion at Time 2, such that the negative relationship between the leader's sleep duration and leader emotional exhaustion would be enhanced under better (versus poorer) leader sleep quality conditions. Results showed that, after adjusting for all other model factors, the link between leader sleep duration at Time 1 and leader emotional weariness at Time 2 was not substantially moderated by leader insomnia symptoms at Time 1 (b = -0.00, SE = 0.02, p = 0.92, 95% CI [-.04,.04]). As a result, Hypothesis 5a was not verified. Results also showed that, after adjusting for all other factors in the model, leader sleep dissatisfaction at Time 1 did not substantially attenuate the connection between leader sleep duration at Time 1 and leader emotional weariness at Time 2 (b = 0.02, SE = 0.02, p = 0.26, 95% CI [-.01,.05]). As a result, Hypothesis 5b was not verified. For an overview of the consequences of moderation, see Table 3.
As stated in hypotheses 6a–b, the indirect impact of leader sleep duration at Time 1 on employee ratings of general supervisor support at Time 3 would be moderated by leader sleep quality at Time 1 (i.e., sleep dissatisfaction [6a] and insomnia symptoms [6b]), via leader emotional exhaustion at Time 2. The moderated mediational (conditional indirect effect) models can be evaluated in accordance with Preacher and colleagues' (2007) recommendations by focusing the moderator at conditional values of interest, calculating model parameters, and interpreting the direct effects as straightforward slopes. In order to compute the interaction terms, the predictor variable (leader sleep duration) and the moderators (leader sleep discontent and leader insomnia symptoms) were centred.
Results showed that, after accounting for all other model variables, leader sleep dissatisfaction at Time 1 had no discernible impact on the indirect impact of leader sleep duration at Time 1 on employee ratings of general supervisor support at Time 3 (conditional indirect effect = -0.00, 95% CI [-.01,.00]). As a result, Hypothesis 6a was not verified. Results also showed that, after accounting for all other model variables, leader emotional exhaustion at Time 2 did not significantly moderate the direct effect of leader sleep duration at Time 1 on employee ratings of general supervisor support at Time 3 (conditional indirect effect = 0.00, 95% CI [-.00,.01]). As a result, Hypothesis 6b was not verified.
According to hypotheses 7a and 7b, leader emotional weariness at Time 2 would attenuate the indirect influence of leader sleep duration at Time 1 on leader (7a) and employee assessments of FSSB at Time 3. Results showed that leader sleep dissatisfaction at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 on leader ratings of FSSB at Time 3 via leader emotional exhaustion at Time 2 (conditional indirect effect = 0.00, 95% CI [-.01,.00]). This is after controlling for all other variables in the model. Results also showed that, after adjusting for all other factors in the model, leader sleep dissatisfaction at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and employee ratings of FSSB at Time 3 via leader emotional exhaustion at Time 2 (conditional indirect effect = -0.00, 95% CI [-.01,.00]). As a result, Hypotheses 7a–b were not verified.
Through leader emotional tiredness at Time 2, hypotheses 7c–d suggested that leader symptoms of insomnia at Time 1 would moderate the indirect influence of leader sleep length at Time 1 on leader ratings of FSSB at Time 3. Results showed that leader insomnia symptoms at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and leader ratings of FSSB at Time 3 via leader emotional exhaustion at Time 2 (conditional indirect effect = 0.00, 95% CI [-.00,.01]). This is after controlling for all other variables in the model. Results also showed that, after adjusting for all other factors in the model, leader emotional weariness at Time 2 did not substantially reduce the indirect effect of leader sleep duration at Time 1 on employee ratings of FSSB at Time 3 (conditional indirect effect = 0.00, 95% CI [-.00,.01]). As a result, Hypotheses 7c–d were not verified.
According to hypotheses 8a–b, leader emotional weariness at Time 2 would help to attenuate the indirect influence of leader sleep duration at Time 1 on leader (8a) and employee ratings of sleep leadership at Time 3. Results showed that leader emotional exhaustion at Time 2 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and leader ratings of sleep leadership at Time 3 (conditional indirect effect = 0.00, 95% CI [-.01,.02]), even after accounting for all other variables in the model.
Results also showed that, after adjusting for all other factors in the model, leader sleep dissatisfaction at Time 1 did not substantially attenuate the indirect effect of leader sleep duration at Time 1 and employee perceptions of sleep leadership at Time 3 via leader emotional tiredness at Time 2. As a result, none of Hypotheses 8a–b were verified.
According to hypotheses 8c-d, leader emotional weariness at time 2 would help to attenuate the indirect influence of leader sleep duration at time 1 on leader (8c) and employee ratings of sleep leadership at time 3. Results showed that leader insomnia symptoms at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and leader ratings of sleep leadership at Time 3 via leader emotional exhaustion at Time 2 (conditional indirect effect = 0.00, 95% CI [-.00,.01]). These findings were made after adjusting for all other variables in the model. The results also showed that, after adjusting for all other model variables, leader emotional exhaustion at Time 2 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and employee ratings of sleep leadership at Time 3 (conditional indirect effect = 0.00, 95% CI [-.01,.01]). As a result, Hypotheses 8c–d was not verified. For a summary of the moderated mediation effects mentioned here, see Table 3.
Post-Hoc Life Satisfaction Analyses
Using the same predictor, moderators, outcomes, and control variables as in my thesis committee's suggestions, I also examined life satisfaction as a mediator inside fully-saturated route models. There are two different theoretical options for studying life satisfaction. First off, emotional tiredness is more indicative of a lack of resources or a loss of resources than than being a resource in and of itself. Contrarily, life satisfaction can serve as yet another resource-based mediator by serving as a proxy for the availability of resources. Second, including life satisfaction in this model adds to the body of knowledge already available and the WNS theoretical model (Crain et al., 2018). The current study's emphasis on positive framing, as discussed in the introduction section, might have been strengthened by including life satisfaction as a mediator. This would have enhanced the paper's original contribution relative to earlier work on sleep and leader behaviour, which has largely been negative. A resource-based mechanism that was implicitly suggested in the WNS framework is life satisfaction. As a result, I investigated life satisfaction as a mediator and the idea of resource availability among leaders in the relationship between leader sleep and subsequent support behaviours.
I checked the validity of the life satisfaction regression assumptions first. The distribution of life satisfaction was normal overall. Eleven univariate outliers were initially analysed, but there was no theoretical justification to exclude the instances from the analysis. Additionally, no multivariate outliers were found after assessing them. Outliers were thus kept around for analysis. According to ICC analysis, there is a significant relationship between life happiness and work group (0.18). Multilevel modelling was thus employed in the analyses. In general, leaders reported feeling content with their lives (M = 3.80, SD = 0.61). It is noteworthy that leader life happiness at Time 2 (r = -0.18, p .01) was significantly and adversely linked with leader sleep dissatisfaction at Time 1 (r = -0.18, p .01). Leader emotional weariness at Time 2 and leader life satisfaction at Time 2 had a significant negative correlation (r = -0.20, p .01). Last but not least, leader assessments of FSSB at Time 3 were significantly and positively correlated with leader life satisfaction at Time 2 (r = 0.20, p .01). Table 1 provides references (i.e., descriptives and correlations).
When considering life happiness as a mediator, there were generally no significant direct effects, indirect effects, moderation effects, or conditional indirect effects, although I do list these findings below. The results are summarised in Tables 5 and 6. First, results showed a non-significant relationship between leader sleep duration at Time 1 and employee-rated general supervisor support at Time 3 through leader life satisfaction at Time 2 (indirect effect = -0.02, 95% CI [-.06,.00]). This was true even after adjusting for all other variables in the model. Results also showed a non-significant connection between leader life satisfaction at Time 2 and leader rated FSSB at Time 3 (indirect effect = 0.01, 95% CI [-.01,.04]). As a result of leader life satisfaction at Time 2, the results showed a non-significant link between leader sleep duration at Time 1 and employee-rated FSSB at Time 3 (indirect effect = -0.02, 95% CI [-.06,.00]). As a result of leader life satisfaction at Time 2, the results showed a non-significant link between leader sleep duration at Time 1 and leader-rated FSSB at Time 3 (indirect effect = 0.01, 95% CI [-.01,.04]). Additionally, findings showed a non-significant link between leader life satisfaction at Time 2 and employee-rated sleep leadership at Time 3 (indirect effect = -0.02, 95% CI [-.07,.01]). Finally, it was discovered that there was no correlation between leader sleep duration at Time 1 and leader-rated sleep leadership at Time 3 (indirect impact = 0.00, 95% CI [-.04,.04]).
The link between leader sleep duration at Time 1 and leader life satisfaction at Time 2 was not substantially moderated, according to the results of the moderation analysis (b = 0.01, SE = 0.01, p = 0.32, 95% CI [-.01,.04]). As a result, the link between leader sleep duration at Time 1 and leader life satisfaction at Time 2 was not substantially moderated by leader insomnia symptoms at Time 1 (b = -0.01, SE = 0.01, p = 0.47, 95% CI [-.03,.01]).
Finally, data showed that leader life satisfaction at Time 2 did not significantly moderate the indirect effect of leader sleep duration at Time 1 (conditional indirect effect= -0.00, 95% CI [-.01,.00]). This was for the larger moderated mediation model. In addition, the results showed that leader symptoms of insomnia at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and employee assessments of overall supervisor support at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = 0.00, 95% CI [.00,.01]). In the case of the FSSB, the findings showed that leader sleep dissatisfaction at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and leader ratings of the FSSB at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = 0.00, 95% CI [-.00,.02]), or employee ratings of the FSSB at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = -0.00, 95% CI [-.01 The results also showed that the indirect effects of leader sleep duration at Time 1 and leader ratings of FSSB at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = -0.00, 95% CI [-.01,.00]) or employee ratings of FSSB at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = 0.00, 95% CI [.00,.01]) were not significantly moderated by leader insomnia symptoms at Time 1 (see results). Finally, the results for sleep leadership showed that leader sleep dissatisfaction at Time 1 did not significantly moderate the indirect effect of leader sleep duration at Time 1 and leader ratings of sleep leadership through leader life satisfaction at Time 2 (indirect effect = 0.00, 95% CI [-.00,.01]) or employee ratings through leader life satisfaction at Time 2 (conditional indirect effect = 0.00, 95% CI [-.01,.01]). The results also showed that the indirect effect of leader sleep duration at Time 1 and leader ratings of sleep leadership at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = 0.00, 95% CI [-.01,.00]) or employee ratings of sleep leadership at Time 3 via leader life satisfaction at Time 2 (conditional indirect effect = 0.00, 95% CI [-.00,.01]) were not significantly moderated by the leader's symptoms of insomnia at Time 1 either.
In this study, I looked at how often and how well people sleep in relation to downstream leader support behaviours, including general supervisor support, FSSB, and sleep leadership. A further hypothesis I made was that these interactions are linked by emotional weariness. These theories matched the WNS theoretical framework (Crain et al., 2018), which contends that sleep affects subsequent work behaviours by producing energy resources. A multilevel route analysis's findings showed that no hypothesis was validated. I present alternate theoretical hypotheses for why the absence of findings may have happened in the sections that follow.
Insights Based on Leader Support
It is crucial to initially draw attention to intriguing descriptive data because they offer original lines of inquiry for subsequent research, even though there were no significant effects in this study and no evidence for hypotheses. In particular, average leader and employee ratings of FSSB (M = 4.11, SD = 0.90; M = 4.10, SD = 0.49, respectively) and employee ratings of general supervisor support (M = 4.23, SD = 0.80) were significantly higher than average employee ratings of sleep leadership (M = 2.51, SD = 0.48) and leader ratings of sleep leadership (M = 2.70, SD = 0.78) were. In organisational research, sleep leadership is a more recent concept (Adler et al., 2021; Gunia et al., 2015; Gunia et al., 2021; Sianoja et al., 2020), and as a result, little study has been done on how leaders and employees perceive sleep leadership. Therefore, it is crucial to discuss sleep leadership in this particular group. These lower sleep leadership values could mean a number of various things.
Leaders may not be practising sleep leadership because they may not feel comfortable bridging the sleep-related nonwork boundary with their staff. Comparing sleep leadership to general supervisor support and FSSB, it may be more difficult for leaders to engage in sleep leadership because both leaders and employees may need to agree on what is appropriate and comfortable to talk about in order for effective sleep leadership to be established by the leader and perceived by the employee. Given the lengthy history of the military undervaluing sleep and the overwhelming incidence of sleep-related problems in the sector, this may be more true in a military setting, where FSSB may be more of an expected type of assistance compared to sleep leadership (e.g., Gordon et al., 2021). As a result, supervisors may need to be more aggressive to engage in sleep leadership than general supervisor support or FSSB. Additionally, there may be less possibilities for sleep leadership to occur spontaneously in the workplace setting. Finally, even though the scale being used does not assess the degree to which employees may request sleep leadership, it is possible that employees dislike receiving support for their sleep because they may perceive it as an invasion of their privacy, particularly because their work leader is stepping over a line that is frequently not discussed in professional settings.
We may gain a unique perspective of how leaders rate themselves on their support behaviours compared to their direct employees by looking at both the leader and employee descriptives. It appears that leaders and workers generally agree on the amount of general, familial, and sleep support that the leader gives at work when comparing the averages and standard deviations for the support outcomes. Contrary to what has previously been shown in the supporting literature, this (e.g., Marescaux et al., 2020). Descriptive data also draw attention to the fact that, contrary to FSSB or general supervisor support, which is acknowledged by both the leader and the employee, sleep leadership was displayed by the leader in the workplace significantly less frequently. Overall, comparing accounts of support behaviours from leaders and employees offered novel early insights into the leader-employee relationship. The incorporation and analysis of data from many sources (e.g., leaders and employees) for sleep leadership and FSSB outcomes underscores the distinctive methodological contribution that the current study brings to the organisational literature, as evidenced by these intriguing descriptive findings.
The Relationship Between Sleep Quantity and Support Behaviors
Regarding the expected non-significant findings, leader sleep duration at Time 1 did not substantially correlate with leader support behaviours as evaluated by employees and leaders at Time 3. (i.e., general supervisor support, FSSB, and sleep leadership). These findings are at odds with the WNS theoretical framework's predictions (Crain et al., 2018), which contend that sleep has a significant influence on our attitudes, behaviours, and states in the work domain. These results also disagree with earlier study (Olsen et al., 2016) that explains the impact of sleep on the demonstration of laissez-faire and transformational leadership styles as well as leader performance ratings (Gaultney, 2014). To further understand how the association between sleep, emotional weariness, and downstream leader behaviours develops over time, this study also incorporated 4-month and 9-month time lags. Cross-sectional or weekly studies on sleep and leadership outcomes have already been conducted (Gaultney, 2014; Olsen et al., 201; Svetieva et al., 2017). As a result, the difference between the findings of the current study and those of previous studies may be attributable to the gaps in measurement times. As a result, when results are analysed over shorter time gaps, such as daily or weekly, they may differ. For instance, compared to typical supervisory behaviours, FSSB and sleep leadership are intrinsically more emotional, necessitating positive affect, empathy, and proactivity to be effective (e.g., Ellis et al., 2022; Crain & Stevens, 2018; Sargent et al., 2020). Previous studies have shown that variations in sleep quality are connected to corresponding variations in affect on a daily basis (e.g., Bouwmans et al., 2017; Sonnentag et al., 2008). So, rather than over the course of months, sleep may have a more significant short-term impact on support behaviours. In light of this, it is possible that the gaps in time between survey events are to blame for the differences between the findings of the current study and earlier studies.
The intriguing levels of sleep health of the National Guard leaders provide another possible explanation for the lack of findings in this study. The leaders in the current sample, in particular, had healthy sleep durations with moderate levels of variance (M = 7.37, SD = 0.99), which may have limited the ability of this study to truly uncover a link between sleep deficiency and subsequent energy and behaviour outcomes because leaders were not experiencing the effects of poor sleep quantity (Hirshkowitz et al., 2015; Ohayon et al., 2017; Watson et al., 2015). Thus, on average, there was no sleep deprivation in the present population, which might have affected the significance of the findings. When we look at each support behaviour separately, other potential alternative explanations become apparent. For instance, in this sample of the National Guard, general supervisor support may be regarded as a basic component of a leader's job description. Because of this, general supervisor support behaviours may come naturally to experienced leaders and so may not call for the same or as much energy activation as other types of behaviours that leaders are expected to engage in at work. High-stakes negotiations, representing the business, interacting with stakeholders, budgeting, performance reviews, safety, or needing to fire staff are a few examples of leadership activities that demand a lot of effort. For instance, research has demonstrated that in addition to actions requiring empathy, such as having to fire employees or resolve conflicts, safety behaviours also call for extra effort (e.g., Zohar & Luria, 2004; Wickens, 2014). (Nowack & Zak, 2020; Cameron et al.,2019). Additionally, depleted leaders are less likely to exhibit transformative leadership and more prone to engage in abusive activities (e.g., Byrne et al., 2014). As a result, leader sleep may be less significant for a leader's ability to perform small, less demanding behaviours, such as general supervisor support, compared to other leader tasks and behaviours. This is because sleep plays a role in refuelling energy and increasing one's tendency to invest effort in behaviours at work.
Emotional Exhaustion as a Mediator
As a result, leader emotional weariness at Time 2 was not a significant mediating factor between leader sleep duration at Time 1 and downstream leader support behaviours at Time 3, according to the data. These findings are at odds with the claims made by the WNS theoretical framework (Crain et al., 2018), which contends that energetic activation mediates the relationship between sleep length and quality and workplace behaviour. These results may also be at odds with claims stated in Quinn and colleagues' (2012) taxonomy of human energy in the workplace, which contends that energetic activation and emotional weariness are closely related. Additionally, sleep has frequently been connected to burnout, of which emotional tiredness is a crucial factor (e.g., Bayes et al., 2021; Ekstedt et al., 2006; Söderström et al., 2012; Toker & Melamed, 2017), indicating that the current findings are also at odds with previous studies. In the limits section that follows, I go through construct validity difficulties as a possible alternate reason for these contradictory results.
Interaction between Sleep Duration and Sleep Quality
The results of the interaction between sleep duration and sleep quality showed that there was no significant interaction between leader emotional tiredness at Time 2 and leader sleep quality and duration at Time 1 as well as a non-significant moderated mediation when looking at the entire model.
These findings run counter to the WNS theoretical framework's assertions that there may be an interaction impact between the quantity and quality of sleep (Crain et al., 2018). The few studies that have so far demonstrated a significant interaction between sleep duration and sleep quality have also reported inconsistent results (Barber et al., 2010; Barnes et al., 2015). The current study and the studies that have indicated a significant interaction differ significantly from one another. The present study specifically looked at the interaction between leader sleep duration and leader sleep quality on leader emotional exhaustion over a 4-month time lag, whereas both studies focused on a daily scale. This suggests that there may be too much time between the measurement occasions for sleep variables and emotional exhaustion, which led to non-significant results. Both studies also looked at other consequences besides emotional weariness. In particular, Barnes and colleagues (2015) discovered the strong interaction with daily ego depletion as the outcome, whereas Barber and colleagues (2010) observed a significant relationship between sleep duration and sleep quality on psychological strain. The lack of statistically significant results in the current study may be explained by the possibility that both outcomes (i.e., psychological strain and daily ego depletion) are more closely associated to sleep-related energy-related outcomes than emotional exhaustion. Last but not least, the samples used in earlier investigations are different from the sample used in this study. In comparison to the National Guard, Barnes and colleagues' (2015) data comes from private and public companies in the United States and Italy, making it more indicative of the typical civilian work.
Additionally, Barber and associates (2010) make use of data from university undergraduate students. Because various samples may exhibit different interactions between sleep duration and sleep quality due to changes in sleep detriments as well as organisational or work factors, the interaction on leader emotional tiredness may not be significant due to sample differences. For instance, unlike undergraduate students at a university, National Guard members may be required to perform potentially dangerous jobs like preparing for domestic situations like natural disasters or using heavy military equipment like aeroplanes or weapons, which may interfere with sleep. Another difference between the National Guard and undergraduate students is that the National Guard may have a more rigid timetable and less conflicting demands when participating in work for the National Guard. For instance, studies have shown that undergraduate students are particularly vulnerable to significant problems with sleep duration and quality because of varying schedules (such as overlapping deadlines, working late hours, and testing structures), competing demands (such as a second job, social obligations, and late-night activities), or financial stressors (such as the cost of living and tuition) (e.g., Gardani et al., 2022).
Unexpected and Non-hypothesized Findings
Despite the fact that none of the study's hypotheses were confirmed, a closer look at the correlation table and every modelled path inside the fully-saturated models revealed some intriguing and surprising results. I report these impacts here rather than in the results section because they weren't hypothesised. For a list of surprising and unanticipated findings, see Table 4. First, results showed significant direct effects between leader ratings of emotional exhaustion at Time 2 and employee ratings of general supervisor support at Time 3, employee ratings of FSSB at Time 3, and employee ratings of sleep leadership at Time 3. This was true even after adjusting for all other variables in the model. This implies that, five months later, employee perceptions of their leader's capacity to provide general, familial, and sleep assistance are strongly and adversely connected to leader emotional weariness. Particularly, employees notice a marked decline in their leader's supportive behaviours when their emotional tiredness level rises. These unexpected findings thus show that a leader's capacity to support workers over time is significantly influenced by their level of emotional weariness. Even more intriguing is the lack of a significant relationship between emotional exhaustion and a leader's perceptions of their own support behaviours. As a result, leaders who are more emotionally exhausted may not believe they support their staff in the same ways as those who are less emotionally exhausted. Emotional exhaustion is strongly aligned with the energetic activation component of both the WNS framework (2018) and Quinn and colleagues' (2012) taxonomy of human energy, even if results did not show potential connection mechanisms between sleep and workplace behaviour via emotional exhaustion. Both theories contend that such energy influences downstream behaviour, and these findings are consistent with both of the theories that were used.
When all other model variables were taken into account, model results also showed two significant interactions. First, there was a significant interaction between leader sleep duration at Time 1 and leader insomnia symptoms at Time 2 on employee ratings of FSSB at Time 3. As a result, under conditions of high leader insomnia symptoms, the relationship between leader sleep duration and employee ratings of FSSB was strengthened. Results also showed that leader insomnia symptoms at Time 1 significantly moderated the relationship between leader sleep duration at Time 1 and employee ratings of sleep leadership at Time 3, such that the positive relationship between leader sleep duration and employee ratings of sleep leadership was strengthened under circumstances of high leader insomnia symptoms. Therefore, leaders who typically get more sleep have a higher likelihood of offering better assistance. However, a supportive leader may be getting enough sleep (i.e., high sleep duration), but still experience high levels of insomnia symptoms because they are obsessing over work-related tasks like meeting their employees' needs in both the familial and sleep domains. As a result, their staff members report that these behaviours have increased. Finally, there was a significant interaction between leader sleep duration at Time 1 and leader sleep dissatisfaction at Time 1 on leader-ratings of sleep leadership at Time 3. In other words, under conditions of high sleep quality and low sleep dissatisfaction, the relationship between leader sleep duration and leader-ratings of sleep leadership was negative and stronger than when sleep dissatisfaction was high. Therefore, regardless of their sleep health, leaders who experience high levels of insomnia symptoms and high levels of sleep duration, or low levels of sleep dissatisfaction and high levels of sleep duration, may be more understanding toward staff members who have obligations outside of work or other sleep-related barriers, and may therefore be more likely to exhibit these support behaviours at work. Additionally, if a leader is experiencing sleep problems, they may be more conscious of how their sleep affects their productivity and may be more likely to promote or encourage sleep in the workplace.
These unexpected and unanticipated findings indicate that employee ratings of both FSSB and sleep leadership, but not employee ratings of general supervisor support, are strongly influenced by the interaction between leader sleep duration and leader insomnia symptoms. In contrast to more emotional, non-work support, like FSSB and sleep leadership, sleep may not be as closely linked to a leader's ability to provide general support. Consequently, the association of leader sleep duration on downstream general supervisor support may be reduced because general supervisor support behaviours may feel generally less taxing and be more automatic regardless of sleep, which further supports why the direct relationship between leader sleep duration and general supervisor support was found to be non-significant.
Overall, the present study answers the WNS authors' requests to investigate the interplay between sleep length and quality and to look into sleep-related connections throughout time (Crain et al., 2018). In addition to a significant interaction between leader sleep duration and leader sleep dissatisfaction, which was significantly linked to downstream leader-ratings of sleep leadership, the results support Crain and colleagues' (2018) WNS framework. Leader sleep duration and leader insomnia symptoms were significantly associated downstream with employee-rated FSSB and employee-rated sleep leadership. Informing future research, the current study identifies how the indirect association between sleep duration and several positive leader behaviours may be strengthened by sleep quality. It also adds more empirical support for this interacting relationship.
In general, this work has repercussions for professionals, groups, and public health initiatives. The unanticipated and non-theorized findings have practical importance despite the fact that the hypothesised associations were shown to be non-significant. Prior studies have demonstrated that practises including refraining from caffeine use or utilising technology right before night, as well as mindfulness activities, can enhance sleep health (e.g., Harvey, 2000; Howell et al., 2010; Mastin et al., 2016; Shallcross et al., 2019). Leaders can utilise this study to support their personal sleep in a proactive manner, but organisations that value leader health and want their leaders to succeed at work should think about offering resources like sleep hygiene and mindfulness trainings. Additionally, the findings showed that employees are more likely to believe a leader is offering FSSB or sleep leadership if they have high sleep duration and high symptoms of insomnia. However, this suggests that leaders may be sacrificing restful sleep for work-related ruminating, which is causing leaders' downstream support behaviours to increase. These strong, non-hypothesized relationships imply that downstream support behaviours displayed by the leader and perceived by the employee both depend on sleep duration and quality in a special and maybe different way.
Therefore, businesses should work to promote sleep-friendly practises and regulations. Organizations might, for instance, take immediate action to remove or eliminate the negative cultural cues that promote the idea that a leader would be more successful if they work harder and sacrifice sleep. Another recommendation is for businesses to set a rigid deadline for work done beyond regular business hours. For instance, establishing a company-wide rule stating that employees are not to check their email after 5 p.m. may enable better segmentation of the work and nonwork domains, giving leaders more time and space to prioritise their sleep health and avoid work-related rumination (e.g., Melo et al., 2021; Sonnentag & Fritz, 2015).
Organizations should give priority to initiatives that lessen the emotional exhaustion that leaders are feeling, like mindfulness-based stress reduction interventions (Hülsheger et al., 2013; McFarland & Hlubocky, 2021), if they want to promote or improve employee perceptions of leader support behaviours. According to research, organisational leaders may be an excellent point of intervention as they straddle an employee's work and nonwork lives (Hammer et al., 2021; Major & Lazun, 2010). There is evidence from a variety of supervisor interventions that educating supervisors to be more understanding of an employee's outside of work life can have a positive impact on an employee's health, well-being, and outcomes connected to their employment (e.g., Brady et al., 2021; Hammer et al., 2011; Hammer et al., 2019; Hammer et al., 2020; Odle-Dusseau et al., 2016; Perry et al., 2020). Non-hypothesized findings, however, showed that leaders who are emotionally spent will not likely be perceived by their direct reports as being supportive of their needs for family time or sleep. Therefore, in order to increase the efficiency of leader-based treatments for encouraging support behaviours like FSSB and sleep leadership, practitioners should work to lessen emotional weariness among leaders.
Finally, public health initiatives can benefit from this study. The link between leader sleep length and downstream employee views of the leader's FSSB and sleep leadership as well as leader perceptions of sleep leadership was shown to be significantly strengthened by components of leader sleep quality. This emphasises how crucial good sleep is for later support behaviours in the workplace. As a result, public health campaigns could change their messaging to emphasise improving sleep quality as well as duration of sleep. For instance, the "7 and up" campaign (American Academy of Sleep Medicine, 2021) could incorporate information or tools related to sleep hygiene (i.e., sleep habits related to sleep quality such as maintaining a consistent sleep schedule or avoiding alcohol or caffeine before bed; Mastin et al., 2006) to help people improve their sleep quality and prevent insomnia symptoms as well. This is in contrast to only promoting a bedtime calculator aimed at improving sleep duration. Additionally, public health campaigns might start assisting educational activities regarding the significance of comprehending both the differences between quantity and quality of sleep.
Limitations and Future Directions
Then I go over a few of the present study's weaknesses. These restrictions include ideas about measurement, theory, generalizability, and study design. Furthermore, I highlight crucial and fascinating future research directions and talk about how to overcome these restrictions.
Study design considerations
The present study's primary limitations include sample size and statistical power.
In multilevel analyses, sample size at each level of the model determines statistical power (Snijders, 2005). Given that there were roughly 175 leaders in the final sample after matching leaders to employees and cleaning all data, it is likely that the sample size at the leader level was insufficient to identify significant connections among the research variables. Kline (2011) suggests 200 instances at the very least for multilevel modelling. Furthermore, Kline (2011) explains how 200 distinct cases can possibly be too few for sophisticated models. The majority of scholars concur that multilevel models and structural equation modelling are "large-sample" studies. Because of the intricacy of this model and the resulting lack of power in this investigation, which may be the cause of the differences between the results and the hypothesised correlations, Another drawback of this study is the time gaps between measurement instances. This study included 4-month and 9-month time lags in accordance with prior recommendations for longitudinal studies on sleep and workplace outcomes (e.g., Crain et al., 2018) to better understand how the relationship between sleep, emotional exhaustion, and downstream leader behaviours develops over time. Future studies should look at how these associations change over shorter time lags, including using day level analysis, even if longitudinal research is valuable. For instance, daily diary studies may show that supervisors reported higher emotional weariness and a decline in positive behaviours the following day on days when they got less sleep and had poorer quality sleep. Therefore, future studies should look into how sleep affects leader results utilising shorter time delays.
The results' generalizability to other samples is another drawback. The military sample from a broader intervention research that was conducted in an effort to enhance sleep and health outcomes served as the basis for the sample used in this investigation. The use of a sample from the National Guard may limit the applicability of the study's findings to broader civilian populations.
Both to less traditional military communities and to both. Despite being full-time workers in a range of jobs, such as human resources or finance/supply, the leaders and employees included in this study nonetheless operate inside the National Guard setting, which may have some differences from non-governmental organisations. Employees of the National Guard, for instance, are regularly trained for high-risk, high-stress circumstances, such as domestic emergencies or anti-drug initiatives, indicating that they may be on call for such events. Additionally, some National Guard jobs include direct interaction with dangerous machinery like planes or weapons, making them safety-sensitive. However, compared to other studies that are solely focused on active-duty soldiers, the sample's generalizability is improved by the variety of employment kinds covered in it. Overall, because leader support is vital regardless of profession, future research should replicate this study across various occupations. Those with unusual schedules (e.g., shiftwork, night work), like those in nursing, the restaurant or hotel business, or even occupations where workers travel frequently, like flight attendants, professional athletes, or construction workers, would be particularly intriguing populations to study.
The role of sleep in downstream leader support behaviours in jobs that have moved to the "front-line" or have a more remote nature, such as first responders, personal care assistants, grocery store employees, or fast-food workers, would also be particularly interesting to investigate given the ongoing global pandemic.
The metrics employed in this study have their own drawbacks. First, the guidelines for the sleep duration and quality scales were different, therefore participants had to whereas participants were asked to evaluate their sleep quality over the previous week, they were asked to indicate their average sleep duration over the previous month. Because participants could more reliably recall their experiences of sleep quality while reflecting on the past seven days as opposed to the past month, this shorter time frame was purposefully chosen for sleep quality. The time frames provided to participants for their subjective assessments of sleep length and sleep quality should be coordinated in future investigations. This study also relies on people's subjective assessments of the quantity and quality of their sleep. Despite the fact that research has shown self-report sleep measures to be valid and reliable, objective measures using actigraph watches to record physiological measurements of activity and rest periods are advised in addition to self-reports to provide a more comprehensive understanding of a person's sleep (e.g., Ganster et al., 2018; Landry et al., 2015).
Using objective measures of sleep length and quality, future study should try to comprehend these postulated links between sleep and the results of leader behaviour. The construct validity and framing of the emotional tiredness measure are similar limitations. When examining the separate components, it becomes clear that the emotional exhaustion measure utilised does not necessarily reflect the primary aspect of burnout (i.e., low energy) as much as it does the leader's capacity for emotionally engaging in others. Although the items are in line with the generally accepted notion of emotional tiredness developed by Shirom and Melamed in 2006, it's possible that this test is actually assessing the leader's emotional interpersonal capability or emotional investment. The current work emphasises sleep's function in energy replenishment and makes use of WNS theoretical It asserts that energetic activation links sleep to subsequent behaviour results (Crain et al., 2018). The emotional fatigue measure, however, was more focused on a result of emotional tiredness's existence or absence than it was on the energy component of burnout and emotional exhaustion, as was originally expected. As a result, emotional tiredness might be acting as a stand-in for emotional energy in this study, and it might be too far removed from the theoretical hypotheses to produce meaningful associations. By taking into account mediators that were capable of more precisely and closely gauging the energy component of emotional weariness, this study may be strengthened. Other mediators that are less proximal and more directly representative of energy, such as energetic activation, ego depletion, or self-regulation, should be taken into account in future studies interested in examining energy's mediating role in the relationship between sleep and downstream behaviours. A further, possibly interesting direction would be to investigate how leader rumination at night may moderate the association between sleep predictors and support behaviour outcomes given the strong interactions of this study. Finally, perceived partner responsiveness, which shows one's assessment of marital resources, or experienced stress, which represents a lack of resources, are additional potentially fascinating mediators. The main hypotheses of this study also focus on determining the degree to which leaders experience emotional exhaustion when interacting with their staff. For example, "I feel I am unable to be sensitive to the needs of coworkers" on this scale's instructions and items asks participants to consider their interactions with "coworkers." These questions do not clearly define who "coworkers" are.
Is referring to, hence the questions did not specifically ask leaders to take into account their relationships with their direct reports. Given that leaders may take into account interactions with other individuals within the organisation, including peers at the same level, their own boss, or their employees, this may have an impact on the study's findings. Different emotional tiredness symptoms may manifest in each of these partnerships. For instance, when asked to think about their contacts with their employees as opposed to their interactions with other leaders, leaders may report feeling more emotional tiredness.
The phrasing of the emotional exhaustion scale items should be matched with the relevant dyadic relationship in future studies (for example, "I feel I am unable to be sensitive to the requirements of my employees").
This study's use of the FSSB short-form measure is another drawback. Although this was a deliberate design decision to prevent participant testing tiredness, it limits our knowledge of the potential connections between sleep and emotional exhaustion and many aspects of FSSB. In order to further understand how sleep may differentially be connected with emotional support, instrumental support, role-modeling, and creative work-family management through the mediator of emotional tiredness, future research should incorporate the entire measure of FSSB. Understanding how the four characteristics of FSSB may be differently impacted by sleep could help inform future treatments meant to promote FSSB in the workplace.
Finally, the validity of the WNS theoretical framework is not fully tested by the current investigation. Future research should try to evaluate the entire theoretical model in order to understand how the links between sleep, emotional weariness, and leader actions may alter when analysed holistically, even if this work pulls from the key concepts proposed by Crain and colleagues (2018). In specifically, the WNS model proposes that sleep precedes and follows attitudes, behaviours, and emotions in the work and nonwork domains.
Despite the fact that the current study only looks at upstream sleep, it would be a particularly intriguing direction for future research to additionally take into account how these anticipated linkages and work behaviour results could affect downstream sleep. Rumination, for instance, can have a significant negative impact on a leader's sleep if they believe they are not supporting their people enough. Although the focus of this study is on actions in the work domain, it is equally crucial to conduct research that explains how leader sleep may affect both their work attitudes and states as well as their nonwork behaviours, attitudes, and emotions. For instance, leader sleep may be connected to feelings of job self-efficacy or even increases in creativity, and these may then be connected to downstream performance in the form of support behaviours.
Additionally, this study did not take into account the impact of physical energy and instead focused on the role of energetic activation as the mediator between sleep and behaviour outcomes. Wearable accelerometer devices should be used by researchers who are interested in examining physical energy as a mediating mechanism.
A computation of oxygen consumption at that rate (Butte et al., 2012; Hills et al., 2014). It's interesting that despite the WNS framework's suggestion that time is a limited resource, this study was unable to examine the impact of perceived time on relationship hypotheses. Future research may find it particularly useful to explore the role of time in the relationships between sleep, emotional tiredness, and leadership behaviour because time is likely to have a significant impact on the relationship between sleep and subsequent work and nonwork outcomes. Finally, given the research on how leaders sacrifice their sleep for work (Ruderman et al., 2017), it is crucial to evaluate how leader sleep may affect relationship satisfaction or work-family conflict because it may be more severe than what non-leader employees experience because of the implicit link between work hours and success among workplace leaders.
The purpose of the current study was to examine the effects of leader sleep on a constellation of downstream support behaviours in the workplace as measured by ratings from leaders and employees. This investigation drew on the work, nonwork, and sleep (WNS) theoretical framework (Crain et al., 2018). (i.e., general supervisor support, FSSB, and sleep leadership). This study also sought to evaluate the relationship between leader sleep duration and sleep quality, as well as the relationship between leader emotional weariness and support actions (i.e., insomnia symptoms and sleep dissatisfaction). Results showed that the correlations that had been predicted were not significant. Unexpectedly, non-hypothesized data indicated a substantial direct and adverse link between leader emotional tiredness and employee-rated support behaviours.
The association between leader sleep length and downstream employee and leader rated support behaviours is reinforced under specific conditions of leader low sleep quality, according to surprise results that also revealed significant interactions. According to these findings, activities that lessen and avoid leader emotional weariness as well as support healthy sleep habits should be given top priority by researchers, practitioners, workplace leaders, and organisations. Public health initiatives should inform the public about the significance of both the quantity and quality of sleep.
Figure 1. Time lagged moderated mediation model of leader sleep duration at Time 1 (i.e., Baseline) on emotional exhaustion at Time 2 (i.e., 4-month), moderated by supervisor sleep quality at Time 1 (i.e., Baseline), on general supervisor support, FSSB, and sleep leadership at Time 3 (i.e., 9-month). Both leaders and their direct employees provided ratings of FSSB and sleep leadership. Control variables (i.e., work schedule, Army vs. Air, child/eldercare responsibilities) not shown.