Work from home and employee well-being: a double-edged sword

Abstract

Background

With the advancement of globalization and technology, remote work has gradually become a feasible and popular work model, particularly with the rapid adoption of working from home driven by the COVID-19 pandemic. This shift not only provides employees with greater flexibility and convenience, potentially enhancing their well-being, but also brings challenges such as blurred boundaries between work and family, and increased social isolation, which can lead to a decline in well-being. Therefore, exploring the dual-edged impact of working from home on employee well-being can help organizations design and implement better policies, enabling employees to balance work and family while improving overall well-being.

Methods

Using the Job Demands-Resources model, this research develops a moderated dual-mediation model to examine the effects of working from home on employee well-being. Data were collected through an anonymous online survey, resulting in a total of 343 valid responses.

Results

Our findings reveal that the direct effect of working from home on employee well-being is not statistically significant, however, its influence is multifaceted. Specifically, working from home can negatively impact well-being by exacerbating family-work conflict, while simultaneously promoting well-being by enhancing job engagement. Furthermore, an individual's confidence in managing work and family responsibilities, referred to as "work-family balance self-efficacy," moderates the relationship between working from home and family-work conflict.

Conclusions

The research findings contribute to advancing theoretical understanding of remote employee management and positive organizational behavior in the digital era. They offer valuable insights for organizations to optimize the management of working from home and support the innovation of human resource management practices in enterprises.

Background

With the rapid development of information technology and the acceleration of globalization, work patterns are undergoing profound transformations. Work from home (WFH), as a flexible work arrangement, is gradually becoming a trend and has rapidly become widespread in specific circumstances [1]. This working mode breaks the temporal and spatial constraints of traditional work environments, offering employees greater autonomy and flexibility while enabling companies to save on office costs. However, the impact of WFH on employee well-being exhibits a double-edged sword characteristic: On one hand, it positively influences mental health and job satisfaction through factors like reduced commuting time [2], improved work-life balance [3], and enhanced personal self-efficacy [4]. On the other hand, prolonged WFH may lead to social isolation [5], blurred professional boundaries [6], and conflicts between family and work roles [7], which can undermine employees'sense of belonging and team collaboration spirit, thereby negatively affecting their well-being. Given that an increasing number of companies are adopting flexible work systems, researching how WFH impacts employee well-being becomes particularly important. Such research can help organizations formulate more humane management policies to promote employee health and productivity. Moreover, in the context of the post-pandemic era, where WFH may become a norm, exploring its complex impact on employee well-being is crucial for preparing society to adapt to future shifts in work patterns and ensuring stable socio-economic development.

WFH refers to the practice of employees completing their work tasks at their residence, representing a specific form of remote work, which more broadly encompasses working from any location outside of the traditional office environment [8]. Existing research largely highlights the positive effects of remote work. For instance, remote work provides employees with greater autonomy by offering increased flexibility in workplace and working hours, which can reduce work-family conflicts and alleviate work-related stress [9]. It also promotes work-life balance, ultimately enhancing employee satisfaction and organizational commitment [10]. Overall, studies indicate a positive association between remote work and outcome variables such as job engagement, employee productivity [11], organizational commitment [12], organizational performance [13], and employee well-being [14]. However, some studies have identified potential negative consequences of remote work. For example, scholars argue that remote work blurs the physical boundaries between work and family, increasing employees’ family responsibilities and thereby intensifying work-family and role conflicts [15]. Additionally, remote workers may experience higher work intensity, greater emotional exhaustion [16], and reduced work enthusiasm and engagement [5, 12]. Some organizations have also reported declines in employee productivity due to remote work, leading them to discontinue offering this option [17].

When examining employee well-being, scholars have conceptualized and defined it within various research frameworks, tailored to specific purposes. However, no consensus has yet been reached on its precise definition or meaning. While there is a general understanding of the concept, a universally accepted definition remains elusive [18]. Based on the diverse definitions and terminologies proposed by scholars, employee well-being is commonly understood as a self-assessed construct reflecting the positive emotions and perceptions employees experience in the workplace. It encompasses work-related factors that significantly shape their overall work experience [19]. Broadly, the determinants of employee well-being are multifaceted and can be categorized into personal factors, organizational factors, and social factors. ① Personal factors. Employees’ self-exploration and experiences within the work environment contribute to the development of psychological functioning and overall well-being [20]. Personal resources such as conscientiousness [21], psychological capital [22], and self-esteem [23] enhance well-being by fostering effective functioning. ② Organizational factors. Work-related demands, such as overtime [24], time pressures [25], and electronic communications from leaders during non-work hours [26], can negatively impact employee well-being. In contrast, factors such as self-efficacy [27], positive workplace relationships [28], and team-building practices [29] have been found to promote well-being. ③ Social factors. Broader social dynamics also influence employee well-being. These include social support [30], the prevalence of information and communication technologies [31], and artificial intelligence [32]. Understanding the interplay of these factors is crucial for developing strategies to enhance employee well-being within organizational contexts.

The relationship between remote work and employee well-being is complex, influenced by both positive and negative factors, as well as various moderating elements. ① Positive factors. Remote work has demonstrated numerous benefits for employee well-being, primarily by enhancing flexibility and autonomy. This work model enables employees to better balance personal and professional responsibilities by eliminating commuting times and allowing autonomous time management. These changes can alleviate life stress and increase overall satisfaction [33]. Additionally, the autonomy afforded by remote work allows individuals to customize their work environments according to personal preferences, fostering job satisfaction, job engagement, and mental well-being [34, 35]. For employees who can minimize distractions and work under favorable home conditions, remote work improves efficiency and focus [36]. This working style reduces task overload and creates an environment conducive to deep thinking, positively influencing psychological well-being. ② Negative factors. Despite its advantages, remote work poses challenges that can negatively impact well-being. A primary concern is the risk of social isolation and loneliness [37]. Limited face-to-face interactions with colleagues may reduce connectedness and diminish the sense of belonging [38]. Another significant issue is the blurring of boundaries between work and personal life. Research shows that remote workers often struggle to maintain a clear separation, leading to prolonged working hours and an increased risk of burnout [39]. Furthermore, the absence of direct supervision and reduced informal communication opportunities may heighten stress and decrease job satisfaction, as employees feel unsupported or uncertain about career advancement [40]. If these challenges remain unaddressed, they can significantly undermine mental health and overall well-being [41]. ③ Moderating factors. The association between remote work and employee well-being is influenced by various moderating factors, including personal preferences, managerial support, and organizational culture. Employees with a strong preference for autonomy generally experience higher satisfaction and well-being in remote work settings [42]. Conversely, individuals who value interpersonal interactions may feel dissatisfied due to the lack of face-to-face engagement [43]. Managerial support is also crucial, compared with remote workers who have limited communication with their supervisors, employees who regularly receive guidance, have open communication channels, and receive recognition from their managers report higher levels of well-being [13, 44]. Additionally, a positive organizational culture, coupled with robust collaboration technologies, helps maintain team cohesion and a sense of belonging, mitigating the negative effects of isolation [45, 46]. Understanding these moderating factors allows organizations to design more effective remote work policies. By addressing diverse employee needs and fostering a supportive work environment, organizations can maximize the benefits of remote work while mitigating its challenges, ultimately enhancing overall employee well-being.

As outlined above, remote work remains a contentious topic. However, there is a notable gap in the literature exploring the dual-edged nature of remote work's impact on employee well-being. WFH, as a distinct form of remote work, presents unique characteristics and challenges, making its relationship with employee well-being particularly worthy of deeper investigation. Unlike shared office spaces or other remote work environments, WFH is often shaped by personal and familial influences, such as household responsibilities, childcare demands, and space constraints. These factors can profoundly affect employees'work experiences and psychological states. Employee well-being is a critical component of a healthy work environment, contributing significantly to organizational performance and competitiveness. Examining WFH’s impact on employee well-being through the lens of the Job Demands-Resources (JD-R) model offers valuable insights. The JD-R model divides factors related to work into job demands and job resources [47]. Job demands refer to negative aspects of work that require sustained effort and can deplete personal resources, whereas job resources encompass positive elements that replenish resources, foster job engagement, and improve outcomes [48]. The dual-pathway hypothesis developed by JD-R model suggests that having sufficient job resources can increase employees'job engagement, leading to positive work outcomes, known as the gain pathway. Conversely, excessive job demands can trigger work burnout, resulting in negative work outcomes, known as the drain pathway [49]. In terms of the gain pathway, when WFH is regarded as a special job resource, an increase in the intensity of WFH will continuously provide individuals with additional resources, thereby increasing their job engagement and enhancing employee well-being. In terms of the drain path, when WFH becomes a specific job demand, the increasing intensity of WFH continuously blurs role boundaries, leading to employees facing work-family conflicts and reducing their overall sense of well-being. WFH has fundamentally transformed how employees perform their tasks, making it essential to distinguish between its roles as a job demand or resource. Recognizing WFH as a double-edged sword underscores the importance of organizational strategies aimed at mitigating its negative effects while amplifying its positive outcomes. Effectively addressing this balance is critical for optimizing employee well-being and enhancing organizational success. Consequently, exploring mechanisms to manage the dual-edged effects of WFH remains a vital area for both academic inquiry and organizational.

In summary, this research aims to investigate the dual-edged effect of WFH on employee well-being through the lens of the JD-R model, focusing on its underlying mechanisms. Job engagement and family-work conflict are proposed as mediators representing the gain and drain pathways, respectively. Additionally, family support and work-family balance self-efficacy are introduced as moderators for the gain and drain pathways. The findings of this research contribute to a deeper understanding of the mechanisms through which WFH impacts employee well-being. By integrating two mediating variables, the study highlights how WFH can function both as a job resource, enhancing job engagement, and as a job demand, intensifying family-work conflict. This dual-perspective approach not only enriches the application of the JD-R model but also deepens its theoretical implications. Furthermore, the incorporation of two moderating variables uncovers the multi-level pathways through which WFH influences well-being, offering valuable insights for comprehensively understanding its complex effects.

Research hypothesis

Drain pathway: the mediating role of family-work conflict

According to the JD-R model, WFH can be considered as a job demand under the influence of certain factors. Firstly, technical barriers. if employees lack the necessary technical support or equipment at home, such as stable internet connections and professional software tools, it can increase the difficulty of completing work tasks, constituting additional job demands [26]. Secondly, communication barriers. WFH may lead to ineffective information flow and reduced communication efficiency among colleagues [50], requiring employees to invest more time and effort in ensuring accurate information transmission, which is another form of increased job demand. Moreover, role ambiguity. WFH blurs the boundaries between work and personal life, making it difficult for employees to distinguish between work time and leisure time [51], leading to a prolonged state of being"always-on,"which increases psychological stress. Lastly, family interference, particularly in households with children or elderly dependents, the needs of family members can frequently disrupt work processes [52], making it challenging for employees to focus on their tasks, thus creating high job demands.

Family-work conflict refers to the interference of roles in the family domain with roles in the work domain, specifically, the general demands, time commitments, and family pressures can interfere with the fulfillment of work responsibilities [53]. When WFH becomes a job demand, it can increase family-work conflict through multiple mechanisms. Specifically, technical barriers and communication barriers can impose additional pressure and challenges on employees as they strive to complete their tasks [34]. These heightened job demands consume employees'time and energy, making it difficult for them to manage their family responsibilities effectively. At the same time, role ambiguity blurs the lines between work and family life. Employees may find themselves handling household tasks during work hours or addressing work responsibilities during family time, leading to poor time allocation and strained family relationships [54]. Furthermore, family interference, especially in households with children or elderly dependents, results in frequent interruptions that hinder employees’ focus and task completion, further increasing psychological pressure [55]. Over time, facing these high job demands can leave employees feeling physically and mentally exhausted, depriving them of the energy needed to meet family members’ needs. This can lead to emotional detachment and inner conflicts. Finally, WFH reduces opportunities for face-to-face interactions with colleagues, leaving employees feeling isolated and unsupported [56]. This lack of workplace social support becomes particularly problematic when addressing family challenges, further intensifying family-work conflict. Here, we propose that:

• H1a: WFH positively affects family-work conflict.

Family-work conflict is an inherent challenge during WFH, affecting both men and women alike. This stems from the family environment, which is often not conducive to supporting full-time learning or working activities for all members, making it easy for individuals to lose focus [57]. WFH also dissolves the physical boundaries between work and family domains [58], allowing work stress to encroach on family life while family responsibilities interfere with work tasks. Consequently, employees struggle to relax after completing their work, leading to blurred psychological and temporal boundaries [12]. These blurred boundaries negatively affect job satisfaction [59] and life satisfaction [60], reducing organizational commitment [61], and causing job burnout [62].

In conclusion, WFH blurs the boundaries between work and family domains, intensifying family-work conflict. This conflict negatively impacts employees'physical and mental health, job satisfaction, and life satisfaction, ultimately leading to job burnout and a diminished sense of well-being. Therefore, the following hypotheses are proposed:

• H1b: Family-work conflict negatively affects employee well-being.

• H1: Family-work conflict mediates the relationship WFH and employee well-being.

Gain pathway: the mediating role of job engagement

According to the JD-R model, WFH provides employees with several valuable job resources. Firstly, time resources. WFH grants employees greater autonomy in managing their time. This flexibility allows employees to adjust their schedules based on personal productivity peaks, family responsibilities, or individual preferences, thereby utilizing time more effectively [45]. Secondly, flexibility resources. The flexibility in work hours and locations afforded by WFH enables employees to arrange their schedules based on personal habits and work preferences. This adaptability promotes a better balance between work and personal life [36]. Thirdly, concentration resources. Reduced workplace distractions create an environment conducive to tasks requiring intense focus [63]. Such an environment enhances productivity and work quality, particularly for employees engaged in deep thinking or handling complex tasks.

Job engagement is a positive, fulfilling work-related state of mind characterized by dedication, energy, and focus [64]. These job resources enhance employees'job engagement through various mechanisms. Specifically, time resources allow employees to have more time for rest and personal interests, reducing fatigue and boosting energy levels. This increased energy enables employees to engage more actively in their work, leading to higher efficiency and creativity [65]. Flexibility resources enable employees to schedule their work hours according to their personal living habits and work preferences, reducing the conflict between work and life. This flexibility enhances employees’ sense of control and autonomy over their work [66]. Such autonomy fosters motivation and responsibility, further increasing job engagement [67]. Concentration resources create a quieter and more focused working environment for employees, which helps improve work efficiency and quality [17]. Completing tasks without interruptions fosters satisfaction and a sense of accomplishment, which, in turn, strengthens their commitment and engagement with work. We propose that:

• H2a: WFH positively affects job engagement.

According to the JD-R model, job engagement, as a critical personal resource, can effectively alleviate work stress and promote a positive psychological state [68]. Greta Mazzetti et al. [69] further revealed that employees with high job engagement often report higher life satisfaction and personal well-being, indicating that job engagement is an essential factor in enhancing employee well-being. Existing studies have shown that job engagement can positively influence employee well-being through various mechanisms: Firstly, job engagement contributes to creating"flow"experiences—when employees are fully immersed in challenging and appealing tasks, they may enter a state of complete absorption, where the sense of time disappears, and attention is entirely focused on the task [70]. This deep work experience not only boosts job efficiency but also brings a strong sense of fulfillment and pleasure, which are core elements constituting employee well-being [71]. Secondly, Jaana-Piia Mäkiniemi et al. [72] found that job engagement not only directly improves employee well-being but also indirectly enhances it by reducing burnout. Highly engaged employees tend to be more resilient against psychological fatigue caused by their profession, maintaining good mental health [73]. Finally, job engagement fosters the formation of a positive organizational culture and social environment. When most members of an organization are highly engaged, teamwork becomes tighter, and innovation and productivity increase accordingly [74]. Such an organizational atmosphere, in turn, further elevates employee well-being, creating a virtuous cycle. Therefore, job engagement not only directly improves the quality of life and job satisfaction of employees but also indirectly enhances their well-being by promoting flow experiences, mitigating burnout, and optimizing the organizational climate.

In summary, for some employees, WFH offers autonomy and flexibility, alleviating work-related stress and enhancing job engagement. This increased engagement motivates them to fully dedicate themselves to their roles, approach their work with passion and optimism, and ultimately experience greater well-being. Therefore, the following hypotheses are proposed:

• H2b: Job engagement positively affects employee well-being.

• H2: Job engagement mediates the relationship between WFH and employee well-being.

The moderating role of work-family balance self-efficacy

Work-family balance self-efficacy refers to employees who strongly believe in maintaining the balance between work and family roles, and are willing to take some actions to avoid family-work conflicts [75]. It plays a crucial role in moderating the impact of WFH on family-work conflict. According to the JD-R model, employees with high work-family balance self-efficacy are more likely to perceive WFH as a supportive job resource rather than an additional job demand. As a result, they are better equipped to manage the challenges of work-family balance, thereby reducing family-work conflict [76]. Specifically, employees with high self-efficacy in balancing work and family responsibilities can create and adhere to clear schedules, effectively allocating time between work and family to prevent work hours from encroaching on family time [77]. They establish distinct boundaries between work and family spaces, minimizing work interference in family life [75]. Additionally, such employees engage in open and honest communication with both family members and colleagues, ensuring mutual understanding and support to address potential conflicts. They are also adept at flexibly responding to unexpected situations in both family and work contexts, quickly adjusting plans and strategies to minimize conflicts arising from unforeseen events [78]. This proactive and flexible approach allows employees with high work-family balance self-efficacy to navigate the complexities of WFH more effectively, fostering a healthier balance between their professional and personal lives.

Conversely, when work-family balance self-efficacy is low, employees struggle to cope with the challenges of WFH, making them more prone to family-work conflict. Employees with low self-efficacy in balancing work and family often face time management difficulties, finding it hard to create and adhere to schedules [3]. This results in an overlap between work and family time. They may also fail to establish clear boundaries between work and family domains, leading to the intermingling of activities and increasing the likelihood of conflict [79]. Insufficient communication is another issue for employees with low work-family balance self-efficacy. They may not engage in adequate dialogue with family members or colleagues, failing to convey or understand mutual needs, leading to misunderstandings and disputes [80]. Furthermore, such employees may find it difficult to quickly adjust plans and strategies, which exacerbates stress and anxiety, further contributing to family-work conflict.

Therefore, work-family balance self-efficacy serves as a crucial moderating factor in the impact of WFH on family-work conflict. Employees with high work-family balance self-efficacy are better equipped to manage the challenges associated with WFH, thereby reducing family-work conflict. In contrast, those with low work-family balance self-efficacy may face a heightened risk of experiencing such conflicts. Based on this, the following hypothesis is proposed:

• H3: Work-family balance self-efficacy moderates the effect of WFH on family-work conflict. Specifically, when the work-family balance self-efficacy is high, the positive effect of WFH on family-work conflict is weakened; Otherwise, it will be enhanced.

The moderating role of family support

Family support is a form of social support that refers to emotional, informational, and material assistance provided by family members [81]. According to the JD-R model, adequate personal resources help employees positively cope with job demands, improving job efficiency and personal well-being [47]. Family support, as a type of personal resource, serves as a buffer against work pressure, assisting employees in better addressing challenges and maintaining high levels of job engagement. Specifically, employees who receive high levels of family support receive emotional encouragement, which helps alleviate work-related stress and boosts motivation [82]. Practical assistance from family members in daily life and household chores reduces the burden on employees, freeing up time and energy to focus on work [83]. Understanding and tolerance from family members also minimize conflicts between family and work, enabling employees to achieve a better balance between the two. These forms of support help employees effectively manage the challenges of WFH, reducing family-work conflicts and amplifying the positive effects of WFH on job engagement [84].

Conversely, employees with low family support lack emotional support, often feeling isolated and stressed, which diminishes their motivation and engagement at work [85]. The limited assistance from family members in daily life and household chores leaves employees feeling time-constrained and drained, making it difficult for them to fully commit to their work [31]. When family members fail to understand the employee's work demands and time commitments, conflicts between family and work increase, leading to heightened stress and anxiety [17]. These add to their workload and pressure, ultimately weakening the positive impact of WFH on job engagement.

In summary, family support plays a critical role in moderating the impact of WFH on job engagement. Employees with high family support are better equipped to handle the challenges of WFH, leading to increased job engagement. Conversely, those with low family support may face greater stress and conflict, resulting in decreased job engagement. Therefore, the following hypothesis is proposed:

• H4: Family support moderates the effect of WFH on job engagement. Specifically, when the level of family support is high, the positive effect of WFH on job engagement is enhanced; Otherwise, it will weaken.

The overall framework is presented in Fig. 1.

Fig. 1.

Theoretical model

Method

Participants

This research targeted employees who had experienced WFH within the past month as the primary study subjects. To ensure sample accuracy, the recruitment process emphasized that participants must have recent WFH experience. A screening question in the survey verified eligibility: participants were asked,"Have you worked from home in the past month?"Respondents who answered"No"were excluded from completing the survey.

An anonymous online questionnaire distribution method was employed, with 381 questionnaires distributed. After excluding invalid responses, 343 valid questionnaires were retained, resulting in an effective rate of 90%. Among the respondents, 45% were male, and 55% were female. The majority were aged 26 to 30 (38%), followed by those aged 31 to 35 (22%). Regarding marital status, 51% were unmarried, and 48% were married. Of the married participants, 9% had no children, 27% had one child, and 13% had two children. In terms of employment sectors, 37% of respondents worked in the private sector, 18% in state-owned enterprises, and 11% in foreign companies. Regarding job roles, 61% were ordinary employees, 17% junior managers, 16% middle managers, and 6% senior managers. Work experience varied, with 29% having more than 10 years of experience, 23% with 6–10 years, and 20% with 3–5 years. Income distribution showed that most participants earned between 5,000 and 10,000 yuan per month, followed by those earning between 10,000 and 15,000 yuan.

Measures

All the scales used in this research are maturity scales. Except for demographic variables, all scales used a 7-point scale, with values ranging from 1 to 7 indicating “strongly disagree” to “strongly agree”. In addition, we use the average variance extraction (AVE) to test the validity of the scale, Cronbach’s α and composite reliability (CR) to test the reliability, and the results are shown in brackets. According to the results, the AVE of each scale are greater than 0.5 [86], the Cronbach’s α and CR are greater than 0.7 [87], indicating both good reliability and validity of the scale.

Work from home was measured in reference to Timothy D. Golden [88] by asking the average number of hours worked from home each week.

Family-work conflict was measured by a 5-item scale compiled by Richard G Netemeyer et al. [53], sample items from this scale include: “The demands of my family or spouse/partner interfere with work-related activities”, “Things I want to do at work don't get done because of the demands of my family or spouse/partner”, “Family-related strain interferes with my ability to perform job-related duties” (Cronbach’s α = 0.933, AVE = 0.738, CR = 0.933).

Job engagement was measured by a 9-item scale compiled by Wilmar B. Schaufeli et al. [64], sample items from this scale include: “At my job, I feel strong and vigorous”, “I feel happy when I am working intensely”, “I get carried away when I am working” (Cronbach’s α = 0.936, AVE = 0.601, CR = 0.930).

Employee well-being was measured by an 18-item scale developed by Xiaoming Zheng et al. [89], sample items from this scale include: “I am close to my dream in most aspects of my life”, “Work is a meaningful experience for me”, “I love having deep conversations with family and friends so that we can better understand each other” (Cronbach’s α = 0.828, AVE = 0.557, CR = 0.957).

Family support was measured by a 3-item scale developed by Gregory D. Zimet et al. [90], sample items from this scale include: “My family really tries to help me”, “I get the emotional help and support l need from my family”, “I can talk about my problems with my family” (Cronbach’s α = 0.914, AVE = 0.547, CR = 0.915).

Work-family balance self-efficacy was measured by a 6-item scale developed by Laurent M. Lapierre et al. [75], sample items from this scale include: “I feel confident that I will succeed in effectively balancing the demands of my work and family life”, “I feel confident that I will schedule my time in such a way that I will have enough time for my work as well as my family life”, “I feel confident that I will succeed in fulfilling my responsibilities both at work and at home” (Cronbach’s α = 0.950, AVE = 0.716, CR = 0.950).

Existing studies indicate that demographic factors such as gender, age, education level, enterprise nature, industry, position, seniority, and income significantly influence employee well-being [91, 92]. Additionally, marital status and the number of children are shown to affect employees'WFH experiences and work-family balance [52]. To minimize the potential impact of these demographic factors on the research findings, this research included them as control variables in the analysis.

Results

Common-method variance test

Common method bias (CMB) refers to the artificial covariation between predictor variables and criterion variables caused by the same data source or raters, the same measurement environment, item context, and characteristics of the items themselves [93]. This artificial covariation can severely confound research results and potentially mislead conclusions, representing a systematic error. To mitigate the potential impact of CMB on the results of this research, we primarily employed the Harman single-factor test, confirmatory factor analysis (CFA), and controlled for the effects of an unmeasured latent methods factor (ULMC). The results of these tests all indicate that there is no significant issue of CMB in this research.

• (1) Harman single-factor test. Harman single-factor test is a widely used method for addressing common method variance with considerable general applicability [94]. This approach involves loading all variables into an exploratory factor analysis and examining the unrotated factor solution to determine the number of factors required to explain the variance [95]. In this research, the test results indicated a KMO value of 0.918, with the Bartlett sphericity test yielding significant results, suggesting that the dataset contained extractable information. The unrotated factor analysis revealed that the largest factor accounted for 30.17% of the total variance, which is below the conventional threshold of 40%. These findings indicate that common method bias was not a significant issue in the sample data used in this research.

• (2) Confirmatory factor analysis (CFA). CFA is used to test whether a set of observed variables can be explained by a single latent variable or construct. This method aims to validate whether the single latent variable proposed in the theoretical model can reasonably represent all observed indicators and assesses the fit between the model and the actual data [96]. According to the results in Table 1, the fit indices of the five-factor model are significantly better than those of the one-factor model, indicating that CMB will not affect the subsequent hypothesis testing and analysis in this research.

• (3) Controlling for the effects of an unmeasured latent methods factor (ULMC). The ULMC method involves using items that are directly affected by a method factor as indicators of this factor. That is, in addition to the original trait factors, all items are also treated as indicators of the method factor, thereby establishing a two-factor model. Generally, if there is a significant difference between the two-factor model and the model containing only trait factors, it indicates a serious CMB [97]. According to the results in Table 1, the changes in fit indices for the ULMC model compared to the pre-control model are all less than 0.03, indicating that the inclusion of the common method factor did not notably improve the model. Therefore, this research does not exhibit significant CMB.

Table 1.

The results of confirmatory factor analysis

Model	${\chi }^2$	$\mathit{df}$	${\chi }^2/df$	$\mathit{TLI}$	CFI	RMSEA	SRMR
ULMC model	1788	946	1.890	0.924	0.934	0.051	0.056
Five-factor model	2147	993	2.163	0.901	0.909	0.058	0.066
Four-factor model	3729	1001	3.726	0.769	0.786	0.089	0.136
Three-factor model	4563	1007	4.531	0.701	0.721	0.102	0.145
Two-factor model	6308	1020	6.184	0.560	0.585	0.123	0.147
One-factor model	6928	1021	6.785	0.509	0.537	0.130	0.148
Criterion			< 3 [98]	> 0.9 [99]	> 0.9 [100]	< 0.08 [101]	< 0.1 [98]

N = 343. Five-factor model: family-work conflict + job engagement + employee well-being + family support + work-family balance self-efficacy. Four-factor model: family-work conflict, job engagement + employee well-being + family support + work-family balance self-efficacy. Three-factor model: family-work conflict, job engagement + family support, employee well-being + work-family balance self-efficacy. Two-factor model: family-work conflict, job engagement + family support, employee well-being, work-family balance self-efficacy. One-factor model: family-work conflict, job engagement, family support, employee well-being + work-family balance self-efficacy

Confirmatory factor analysis

To assess the structural validity of each construct, CFA was conducted using Amos 26 software, with the results summarized in Table 1. As WFH was measured using a single-item scale, it was excluded from the CFA model. The analysis included five factors: family-work conflict, job engagement, employee well-being, family support, and work-family balance self-efficacy. Among the various models tested, including the five-factor model and the one-factor model, the combination with the best results was reported. The findings revealed that the five-factor model demonstrated the best fit, indicating that the five variables in this study possess strong structural validity.

Correlation analysis

Table 2 presents the descriptive statistics, including the mean, standard deviation, and correlation coefficients for each variable. The correlation analysis reveals the following insights: ① The average number of hours spent on WFH in the sample is 9.95. ② Among control variables, WFH shows a significant correlation with industry (r = 0.15, p < 0.01), indicating notable variations in WFH practices across industries. ③ Control variables such as age (r = −0.18, p < 0.01), marital status (r = −0.12, p < 0.05), number of children (r = −0.15, p < 0.05), and seniority (r = −0.11, p < 0.05) are negatively correlated with employee well-being. ➃ Education level (r = 0.12, p < 0.05) demonstrates a positive correlation with employee well-being. ➄ Variables including family-work conflict (r = 0.18, p < 0.01), job engagement (r = 0.17, p < 0.01), family support (r = 0.13, p < 0.05) and work-family balance self-efficacy (r = 0.18, p < 0.01) are significantly positively correlated with WFH. ➅ A significant positive correlation exists between job engagement and employee well-being. (r = 0.21, p < 0.01). ➆ Family-work conflict is positively correlated with job engagement (r = 0.14, p < 0.01). These results provide preliminary support for hypotheses H1a, H1b, H2a, and H2b. Furthermore, as control variables such as gender, nature of work, industry, position, and income are not significantly related to employee well-being, these variables will be excluded from subsequent data analyses to preserve statistical power and degrees of freedom.

Table 2.

Descriptive statistics and correlation analysis results

Variables	1. Gender	2. Age	3. Education	4. Marital status	5. Children	6. Enterprise nature	7. Industry	8. Position	9. Seniority	10. Income	11. WFH	12. Employee well-being	13. Family-work conflict	14. Job engagement	15. Family support	16. Work-family balance self-efficacy
1. Gender	1
2. Age	−0.06	1
3. Education	0.04	0	1
4. Marital status	−0.01	0.56**	−0.08	1
5. Children	−0.03	0.62**	−0.13*	0.94**	1
6. Enterprise nature	0.05	−0.15**	−0.26**	−0.14*	−0.12*	1
7. Industry	0.07	−0.12*	0.04	−0.04	−0.02	−0.08	1
8. Position	−0.06	0.38**	0.03	0.18**	0.20**	0.07	−0.22**	1
9. Seniority	−0.05	0.78**	−0.05	0.50**	0.53**	−0.12*	−0.17**	0.49**	1
10. Income	0.00	0.34**	0.35**	0.26**	0.22**	−0.08	−0.19**	0.37**	0.43**	1
11. WFH	−0.02	−0.04	0.06	−0.10	−0.08	−0.03	0.15**	−0.02	−0.09	0	1
12. Employee well-being	−0.05	−0.18**	0.12*	−0.12*	−0.15*	−0.05	0.05	−0.01	−0.11*	0.04	0.06	1
13. Family-work conflict	−0.03	0.05	−0.17**	0.12*	0.14*	0.02	0.01	0	−0.03	−0.08	0.18**	−0.32**	1
14. Job engagement	−0.12*	0.11*	−0.19**	0.11	0.14*	0.07	0.02	0.06	0.02	−0.09	0.17**	0.21**	0.14**	1
15. Family support	0.01	−0.04	−0.04	0.06	0.05	−0.05	0.12*	0.01	−0.04	−0.05	0.13*	0.35**	0.05	0.55**	1
16. Work-family balance self-efficacy	−0.04	−0.05	0.04	−0.02	−0.01	−0.01	0.03	0.03	−0.04	0	0.18**	0.44**	−0.11*	0.58**	0.62**	1
M	1.55	2.95	2.17	1.50	1.03	5.22	9.95	1.67	3.38	2.77	9.95	4.63	3.49	3.94	4.82	4.89
SD	0.50	1.69	0.87	0.53	1.17	2.33	5.40	0.95	1.41	1.36	5.24	0.83	1.47	1.37	1.17	1.31

M Stands for mean, SD Stands for standard deviation

N = 343

^* stands for p < 0.05

^** stands for p < 0.01

Hypothesis analysis

To examine the relationship between WFH and employee well-being, we employed PROCESS V4.1 (model 7). The analysis utilized the Bootstrap method with 5,000 resamples, and the confidence level was set at 95%. During the mediation analysis, control variables such as age, education level, marital status, number of children, and seniority were included to account for their potential influence. The results of the mediation effect tests are detailed in Tables 3, 4, and 5. These findings provide insights into the underlying mechanisms through which WFH impacts employee well-being, while accounting for the moderating and mediating variables specified in the research model.

Table 3.

Analysis results of mediation test

Regression equation (N = 343)		Fitting index			Significance of regression coefficient
Outcome variable	Predictor variable	R	R2	F	B	t
Family-work conflict	WFH	0.329	0.108	6.780***	0.057	3.909***
	Age				0.088	1.310
	Education				−0.276	−3.113**
	Marital status				0.337	1.115
	Children				0.098	0.671
	Seniority				−0.216	−2.867**
Job engagement	WFH	0.334	0.112	6.004***	0.046	3.335***
	Family-work conflict				0.037	0.732
	Age				0.156	2.497*
	Education				−0.245	−2.940**
	Marital status				0.069	0.245
	Children				0.124	0.913
	Seniority				−0.201	−2.848**
Employee well-being	WFH	0.447	0.200	10.412***	0.010	1.250
	Family-work conflict				−0.197	−6.707***
	Job engagement				0.169	5.344***
	Age				−0.090	−2.466*
	Education				0.044	0.903
	Marital status				−0.079	−0.484
	Children				0.039	0.497
	Seniority				0.020	0.475

^* stands for p < 0.05

^** stands for p < 0.01

^*** stands for p < 0.001

Table 4.

Path coefficient of mediating effect

Path	Path coefficient	SE	p	95% CI
WFH → family-work conflict	0.057***	0.015	0	[0.028, 0.086]
WFH → job engagement	0.046***	0.014	0.001	[0.019, 0.073]
family-work conflict → employee well-being	−0.197***	0.029	0	[−0.255, −0.139]
job engagement → employee well-being	0.169***	0.032	0	[0.107, 0.232]
family-work conflict → job engagement	0.037	0.051	0.464	[−0.062, 0.137]

^*** stands for p < 0.001

Table 5.

Results of chain mediation effect analysis

Path	Value	SE	95% CI
Ind1: WFH → family-work conflict → employee well-being	−0.071	0.023	[−0.121, −0.031]
Ind2: WFH → job engagement → employee well-being	0.049	0.017	[0.021, 0.088]
Ind1-Ind2	−0.120	0.032	[−0.188, −0.066]

The results indicate that WFH is significantly positively associated with family-work conflict, with a path coefficient of 0.057 (p < 0.001). Additionally, a significant negative relationship exists between family-work conflict and employee well-being, with a path coefficient of −0.197 (p < 0.001). The mediation effect value of WFH → family-work conflict→ employee well-being (Ind1) is −0.071, 95% CI: [−0.121, −0.031]. Since the confidence interval does not include zero, the mediating effect is statistically significant. This finding confirms that family-work conflict serves as a mediator in the relationship between WFH and employee well-being.

The results presented in Table 4 demonstrate that WFH is significantly positively associated with job engagement, with a path coefficient of 0.046 (p < 0.001). Furthermore, a significant positive relationship exists between job engagement and employee well-being, with a path coefficient of 0.169 (p < 0.001). The mediation analysis indicates that the indirect effect of WFH → job engagement→ employee well-being (Ind2) is 0.049, 95% CI: [0.021, 0.088]. Since the confidence interval does not include zero, the mediating effect is statistically significant. This finding confirms that job engagement mediates the relationship between WFH and employee well-being.

It is worth noting that the mediating effect sizes shown in Table 4 are all small (< 0.2) but statistically significant (p < 0.001). According to Jacob Cohen [102], the reliability of sample estimates depends significantly on sample size; larger sample sizes provide greater precision under similar conditions. With a substantial sample size (N = 381 > 200), this research achieves significant mediating effects even under stringent thresholds (p < 0.001). Therefore, despite the modest path coefficients, the findings retain theoretical significance. Based on these results, hypotheses H1, H1a, H1b, H2, H2a, and H2b are all supported.

As shown in Table 5, the effect value of path Ind1-Ind2 is −0.120, and the 95% CI is [−0.188, −0.066], which means that the mediating effect of path Ind1 is smaller than that of path Ind2. In other words, the mediating effect of job engagement on the relationship between WFH and employee well-being is greater than that of family-work conflict.

Using PROCESS V4.1, we tested the moderation effect, applying the Bootstrap method with 5,000 repeated samples and a 95% confidence level. Control variables such as age, education, marital status, number of children, and seniority were included in the analysis. The results, as presented in Table 6 and Fig. 2, demonstrate that family-work conflict and the interaction term WFH × work-family balance self-efficacy are significantly negatively correlated (B = −0.034, p < 0.05), 95% CI: [−0.056, −0.012]. Since the confidence interval does not include zero, this indicates that work-family balance self-efficacy significantly moderates the relationship between WFH and family-work conflict. Thus, hypothesis H3 is supported.

Table 6.

Results of moderating effect analysis

Regression model (N = 343)		Fitting index			Significance of regression coefficient		95% CI
Outcome variable	Predictor variable	R	R2	F	B	t
Family-work conflict	WFH	0.394	0.155	7.681 ***	0.071	4.850***	[0.042, 0.099]
	Work-family balance self-efficacy				−0.204	−3.492***	[−0.320, −0.089]
	WFH × work-family balance self-efficacy				−0.034	−3.033**	[−0.056, −0.012]
	Age				0.115	1.734	[−0.015, 0.245]
	Education				−0.288	−3.314***	[−0.460, −0.117]
	Marital status				0.360	1.218	[−0.221, 0.941]
	Children				0.087	0.605	[−0.195, 0.368]
	Seniority				−0.228	−3.090**	[−0.374, −0.083]
Job engagement	WFH	0.614	0.377	25.229 ***	0.030	2.655**	[0.008, 0.053]
	Family support				0.609	11.766***	[0.507, 0.710]
	WFH × family support				−0.008	−0.892	[−0.027, 0.010]
	Age				0.179	3.415***	[0.076, 0.283]
	Education				−0.253	−3.627***	[−0.390, −0.116]
	Marital status				−0.063	−0.268	[−0.527, 0.400]
	Children				0.097	0.849	[−0.127, 0.321]
	Seniority				−0.159	−2.697**	[−0.275, −0.043]

^* stands for p < 0.05

^*** stands for p < 0.001

Fig. 2.

The moderating effect of work-family balance self-efficacy on the relationship between WFH and family-work conflict

To visualize the moderating effect of work-family balance self-efficacy, we select the values of WFH and family-work conflict variables under the work-family balance self-efficacy at a high level (M + SD) as well as the work-family balance self-efficacy at low level (M-SD), respectively, and plot the interaction effects of the moderating variables. In Fig. 2, the slope of low work-family balance self-efficacy is 0.115 (p < 0.001), 95% CI: [0.072, 0.156], the slope of high work-family balance self-efficacy is 0.027 (p = 0.165), 95% CI: [−0.011, 0.064]. The results show that when employees'work-family balance self-efficacy is low, WFH has a significant positive effect on family-work conflict, when employees'work-family balance self-efficacy is high, the effect of WFH on family-work conflict is not significant.

The results presented in Table 6 indicate that job engagement is negatively correlated with the interaction term WFH × family support (B = −0.008). However, this relationship is not statistically significant, as the 95% CI is [−0.027, 0.010] includes zero. This lack of significance suggests that family support does not moderate the relationship between WFH and job engagement. Consequently, Hypothesis H4 is not supported. These findings imply that, while family support is often considered a beneficial resource, it does not play a significant role in influencing the impact of WFH on job engagement within this study's sample.

In conclusion, the results of the hypothesis analysis are shown in Fig. 3.

Fig. 3.

The result of the theoretical model

Discussion

Based on the JD-R model, we constructed a theoretical model to explore the double-edged sword effect of WFH on employee well-being. The main conclusions are as follows:

• (1) This research verifies the double-edged sword effect of WFH on employee well-being, and it is a complete mediating effect. The direct impact of WFH on employee well-being is not significant, but WFH can weaken employee well-being by inducing family-work conflict and enhance employee well-being by increasing job engagement. This coincides with the contradictory view that WFH has its pros and cons [50]. On the one hand, we confirm that WFH increases the permeability of boundaries between work and family domains, exacerbating family-work conflict and negatively impacting employee well-being, which is contrary to the research conclusion that"remote work can increase boundary flexibility and reduce work-family conflict [103]". On the other hand, we confirm that WFH can promote job engagement, thereby enhancing employee well-being. This is consistent with previous research findings [104, 105]. Employees of this kind can maintain boundary permeability and integrate their work and family roles, effectively avoiding role conflicts and fully immersing themselves in their work roles within a given time, which has a positive impact on their well-being. Furthermore, we reveal that job engagement plays a greater mediating role between WFH and employee well-being than the mediating role of family-work conflict. This indicates that the positive effect of WFH on employee well-being is greater than the negative effect, and promoting job engagement is more effective in enhancing employee well-being than reducing family-work conflict. Since job engagement is a subjective factor for employees, and family-work conflict is an objective factor in the employees'working environment, existing research suggests that the impact of subjective factors on employee well-being is greater than the impact of objective factors [106].

• (2) Work-family balance self-efficacy can regulate the relationship between WFH and family-work conflict. Specifically, when employees have a high self-efficacy in work-family balance, the positive impact of WFH on family-work conflict is not significant; when employees have a low self-efficacy in work-family balance, the positive impact of WFH on family-work conflict is significant. This research conclusion indicates that when employees believe they have the ability to balance their work and family roles, we can predict that these employees are able to avoid or mitigate the impact of family-work conflict in both voluntary and involuntary WFH situations. When employees have low self-efficacy in work-family balance, they will find it difficult to avoid various forms of family-work conflict when working remotely. To some extent, the forced integration of work and family roles makes it harder to maintain role boundaries. Therefore, when employees with low self-efficacy in work-family balance are forced to work remotely, it is harder for them to maintain role boundaries and, in turn, experience more family-work conflict.

• (3) We fail to confirm the moderating effect of family support between WFH and job engagement, which may be due to the following reasons: Firstly, job engagement is mainly influenced by work-related factors and employees'personal resources [107]. Since family support is neither considered a work resource nor a personal resource, its moderating effect is not significant. Secondly, the impact of WFH on employees largely depends on how organizations respond to the situation, especially in terms of technical support for WFH, assistance in setting up home facilities and providing timely information access. Therefore, the moderating effect of organizational support may be more significant than that of family support. Thirdly, the level of job engagement is not at a stable level, and its degree of change is not consistent [17]. Although this research investigated the WFH situation in the past week, employees'perception of family support may be a long-term accumulation that does not match the level of job engagement in the past week. Therefore, there are some measurement errors leading to the non-significance of the moderating effect.

Theoretical implications

Firstly, although the JD-R model has been extensively applied to explore how job demands and job resources influence job engagement [68], job satisfaction [108], and well-being [73], most studies have focused primarily on traditional office environments [29, 109]. This research incorporates the emerging work mode of WFH into the analysis, unveiling its dual role as both a job resource and job demand. This dual perspective provides a new lens for understanding employee well-being in WFH environments. This not only broadens the application context of the JD-R model but also adds new dimensions to its theoretical framework, thereby enabling the model to explain the complex dynamics present in modern work settings more comprehensively.

Secondly, existing research often examines the impact of WFH from a singular perspective [14, 42], such as emphasizing either its flexibility [17] or challenges [6]. However, this research introduces family-work conflict and job engagement as mediating variables, revealing a multi-level influence pathway for WFH on employee well-being. This not only deepens our understanding of the mechanisms underlying the formation of employee well-being but also demonstrates the double-edged sword effect of WFH—it can enhance well-being through positive pathways (e.g., increasing job engagement) but may also weaken it through negative pathways (e.g., exacerbating family-work conflict). Such findings provide a more nuanced analytical framework for future research.

Thirdly, while existing research have identified several factors that may influence employee well-being, such as human resource management systems [110], leadership [111], and job crafting [112], they often overlook the role of moderating factors [113]. This research fills this theoretical gap by introducing work-family balance self-efficacy and family support as moderating variables. The inclusion of these moderators not only explains why similar workplace environments can have differing impacts on the well-being of different employees but also proposes an integrative theoretical framework that combines factors at the individual level with those at the social support level. This aids in more accurately predicting and explaining individual differences in performance within WFH environments, providing managers with more effective strategic recommendations to enhance employee well-being and job efficiency.

Practical implications

Based on the above research findings, we offer the following management practice recommendations, aimed at helping organizations better balance the double-edged impact of WFH policies on employee well-being.

Firstly, organizations should recognize that WFH has the potential to both weaken employee well-being by increasing family-work conflict and strengthen it by enhancing job engagement. Therefore, management needs to take measures to reduce the risk of family-work conflict while actively promoting job engagement. For instance, offering flexible work hours can allow employees to adjust their schedules according to their personal rhythms. Establishing clear boundaries between work and family life by encouraging employees to set up dedicated workspaces and time slots can help minimize the likelihood of role conflicts. Additionally, organizations can foster job engagement through recognition and reward systems or by regularly organizing online team-building activities to reinforce a sense of belonging and responsibility among employees, thereby enhancing their overall well-being.

Secondly, to mitigate family-work conflict caused by WFH, particularly for employees with lower work-family balance self-efficacy, organizations can offer targeted support and services. For example, they can conduct training courses or workshops that teach employees how to effectively manage the balance between work and family life; provide counselling services to assist employees in resolving specific family-work conflict issues; and foster a supportive corporate culture where employees feel that their needs are understood and valued. These initiatives can help enhance employees'work-family balance self-efficacy, giving them greater confidence to address the challenges posed by WFH.

Thirdly, although this research indicates that family support does not significantly moderate the relationship between WFH and job engagement, this does not mean that the influence of the family environment on employees can be overlooked. On the contrary, organizations should recognize the importance of family support and seek ways to collaborate with families. For example, through communication channels, organizations can inform family members about the WFH support and resources offered by the company, creating a home environment that is conducive to employees'work. At the same time, organizations should strengthen their support systems, particularly in terms of technical support, home office setup, and information access, ensuring that employees can smoothly adapt to the WFH model. This not only improves the employee work experience but also indirectly promotes family harmony, thereby enhancing overall employee well-being.

Finally, considering the instability and variability of job engagement levels, organizations may consider adopting more long-term data collection methods to more accurately assess the impact of WFH on employees. By continuously tracking employees'job engagement, organizations can gain a more comprehensive understanding and adjust their WFH policies accordingly to better meet the actual needs and evolving trends of employees. Additionally, leveraging technological tools for real-time feedback and adjustments—such as using apps to monitor workloads and provide recommendations—can help maintain high levels of job engagement, thereby enhancing employee well-being.

Specifically, this research explores both the drain pathway and the gain pathway through which WFH affects employee well-being. Therefore, when implementing or optimizing WFH policies, organizations should adopt a dual-track strategy: the primary goal is to mitigate the negative impacts of the drain pathway—family-work conflict—by enhancing employees'self-efficacy in achieving work-family balance. At the same time, efforts should be focused on strengthening the positive effects of the gain pathway—job engagement—by unleashing the autonomy and concentration potential associated with WFH, thereby maximizing its positive impact on employee well-being.

• (1) To mitigate the drain pathway, organizations should implement multi-level interventions to enhance employees'work-family balance self-efficacy. First, systematic capacity-building programs—such as time management training, psychological detachment workshops, and digital tool application guidance—can help employees master practical methods for setting boundaries between work and family. Second, personalized support should be provided to employees with special needs, including flexible scheduling and emergency care subsidies as tangible assistance. More importantly, a robust psychological support system should be established, offering specialized skill training and professional counselling through employee assistance programs to continuously strengthen employees'confidence and ability to cope with family-work conflict. The overarching goal of these measures is to help employees develop clear boundary management strategies, thereby effectively reducing the likelihood and impact of family-work conflict.

• (2) To strengthen the gain pathway, organizations need to prioritize creating a WFH environment that enhances job engagement. Technologically, they should actively promote asynchronous collaboration tools and virtual co-creation platforms, minimizing unnecessary synchronous meetings to provide employees with more opportunities for deep work. Culturally, a robust remote recognition system should be established, using real-time feedback and informal interactions to counteract the emotional detachment caused by physical distance. Notably, incentive mechanisms should be redesigned to accommodate the unique aspects of WFH, incorporating health management and work equipment support to improve employees'overall work experience. The essence of these strategies lies in fully leveraging the flexibility of WFH. By optimizing job design and organizational support, companies can continuously stimulate employees'intrinsic motivation and enthusiasm, ultimately achieving a dual enhancement of well-being and productivity.

Limitations and future directions

This research has the following limitations:

Firstly, this research does not examine the moderating effect of family support on the relationship between WFH and job engagement, which may be attributed to the timeliness of the survey data. The data were collected at a single point in time and from a single source, limiting our ability to capture the dynamic relationships among variables. Future research could employ a multi-wave approach to further explore the mediating and moderating effects in the relationship between WFH and employee well-being. Utilizing longitudinal designs and collecting data across multiple time points would provide a deeper understanding of the dynamic interactions between variables, validate causal relationships, and help mitigate potential confounding factors.

Secondly, this research focuses solely on internal factors, such as family and personal variables, while neglecting the influence of external factors, including the work environment, company policies, and national guidelines. Additionally, the research method employed presents certain limitations. Future studies should explore the mechanisms through which external factors impact the relationship between WFH and employee well-being. A comprehensive approach, incorporating various research methods, will provide a more holistic understanding of this relationship.

Thirdly, this research is limited by its focus on the specific context of WFH in China, which may not fully capture the broader dynamics of WFH in other regions. Furthermore, the measurement of WFH is based solely on the number of hours worked, which fails to account for the qualitative aspects of WFH. Work patterns and systems can vary significantly across cultures, and future research could explore WFH in different countries and in more diverse work environments. This would enhance the generalizability and applicability of the research framework.

Conclusion

Drawing on the JD-R model, this research investigates the relationship between WFH arrangements and employee well-being, providing valuable theoretical and practical implications for the evolving nature of work. While WFH offers both advantages and challenges, organizations should aim to maximize its benefits and minimize its drawbacks by implementing effective management and supervision strategies to support employee well-being in WFH settings. Meanwhile, employees must adapt to technological advancements and the changing demands of the modern workplace, proactively maintaining their well-being across diverse work environments.

Abbreviations

WFH

Work from Home

JD-R

Job Demands-Resources

AVE

Average Variance Extraction

CR

Composite Reliability

CMB

Common Method Bias

CFA

Confirmatory Factor Analysis

ULMC

Unmeasured Latent Methods Factor

KMO

Kaiser–Meyer–Olkin

TLI

Tucker-Lewis Index

CFI

Comparative Fit Index

RMSEA

Root Mean Square Error of Approximation

SRMR

Standardized Root Mean Squared Residual

M

Mean

SD

Standard Deviation

CI

Confidence Interval

SE

Standard Error

Authors’ contributions

Conceptualization, J.D.; methodology, J.D., Z.T., and Y.Z.; software, J.D.; validation, J.D. and Y.S.; formal analysis, J.D.; investigation, J.D. and Y.H.; resources, J.D.; data curation, J.D.; writ-ing—original draft preparation, J.D.; writing—review and editing, J.D.; visualization, J.D.; su-pervision, J.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data availability

The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All procedures in the study were approved and all methods and experimental protocols were performed according to the guidelines of the Declaration of Helsinki. This study involves human participants and was approved by the Ethics Committee of Logistics and E-commerce College, Zhejiang Wanli University (No. ZWU-IRB-20235001). Informed consent was obtained from all subjects involved in the study. This study was explained to the interviewees and subjects before the study, their consent was obtained, and the interviewees and subjects gave their consent for their data to be used as research material.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.