Leader–member exchange and emotional exhaustion in registered construction engineers: mediating role of active management by exception and moderating role of self-esteem

The construction engineer qualification system originated in the United Kingdom. Based on extensive international experience, China began implementing the construction engineer qualification system in 2003 and, since 2008, has mandated that project leaders for construction projects must be registered construction engineers. A registered construction engineer is a professional who has passed the necessary examinations or assessments to obtain the People’s Republic of China Construction Engineer Qualification Certificate and has registered according to regulations to obtain the People’s Republic of China Construction Engineer Registration Certificate. These professionals serve as project leaders, project technical leaders, and engage in related activities for construction units. Registered construction engineers play a significant role in the continuous development of China’s construction industry and economy. By the end of 2022, there were 143,621 construction enterprises nationwide, employing 51.84 million people, and achieving a total construction output value of 31,197.984 billion yuan, up 6.45% year-on-year, with an added value of 8,338.31 billion yuan, accounting for 6.89% of the GDP (Zhao et al., 2023). However, according to data from the China Construction Engineer Network, as of the end of 2022, there were only 587,674 registered first-class construction engineers nationwide, and this number is declining (China Construction Engineer Network, 2023). With the implementation of the “Belt and Road” initiative and new urbanization, the business of Chinese construction enterprises both domestically and overseas is rapidly increasing. Against this backdrop, the sharp increase in demand for talent in the construction market, coupled with the scarcity of high-quality professionals, has become increasingly prominent issues. The sustainable development of human resources in the construction industry has become a significant constraint on the industry’s sustainable development.

The construction industry is characterized by a relatively harsh working environment, with non-fixed work locations, outdoor and high-altitude operations, and many technically complex and difficult tasks (Leung et al., 2008). Therefore, the construction industry is a high-stress industry, and its practitioners are a high-risk group for occupational psychological health problems (Bowen et al., 2014). Compared to general construction workers, construction engineers are scarce high-end professionals who require more academic attention. Previous research has mostly focused on the professional capabilities and professional integrity of registered construction engineers, without giving enough attention to their working conditions and experiences (Lingard and Francis, 2004). Moreover, scholars have paid insufficient attention to the occupational psychological health of construction workers and the factors influencing their occupational psychological stress (Enshassi et al., 2015). Yang et al. (2017) noted that in China, the government, the construction industry, and enterprises do not pay enough attention to monitoring and intervening in work stress. Based on this, this study attempts to investigate the issue of emotional exhaustion - defined as a state of physical and emotional depletion resulting from prolonged exposure to high job demands and continuous work stress -among Chinese registered construction engineers from a humanistic perspective.

According to the Job Demands-Resources (JD-R) model proposed by Bakker and Demerouti, factors influencing work can be categorized into job demands and job resources (Bakker and Demerouti, 2007). Job demands are negative factors requiring continuous physical, mental, and emotional effort, such as work overload, job insecurity, and role conflict. These negative factors can cause energy depletion (Schaufeli and Bakker, 2004). Job resources are positive factors that help achieve work goals, reduce physiological and psychological costs, and promote individual growth and development, such as leadership support, colleague support, job autonomy, and performance feedback. These resources can be physiological, psychological, social, or organizational (Demerouti et al., 2001). From the depletion path perspective, the work process is also an energy-consuming process. As energy depletes, individual stress increases, and physical and mental fatigue intensifies. When a large amount of resources invested do not yield the expected return, employees may experience emotional exhaustion symptoms (Maslach et al., 2001). Individuals experiencing emotional exhaustion are in a state of extreme fatigue, with emotional resources excessively depleted, leading to exhaustion and loss of energy. In an emotionally exhausted state, individuals may experience anxiety, tension, depression, and other negative emotions, adversely affecting their physical and mental health and work performance, reducing work commitment and dedication, and even leading to counterproductive behaviors that can cause major safety accidents. From the gain path perspective, job resources are inherently motivational and can stimulate employees' work motivation. Abundant job resources can effectively suppress job burnout, increase work engagement, and produce positive effects (Hakanen et al., 2006).

The construction industry is a long-term high-stress, high-demand industry, and its practitioners are highly prone to job burnout. Yang et al. (2017) believe that Chinese construction project managers are under high levels of work pressure for extended periods, at high risk of job burnout. Additionally, due to the industry’s inherent characteristics and work hazards, the construction industry has always emphasized safety and quality, famously encapsulated in the saying “Safety first, a century-old plan.” Besides necessary external protective measures, attention to the psychological state of construction workers is imperative, especially the negative psychological states that might lead to safety accidents, such as job burnout characterized primarily by emotional exhaustion (Leung et al., 2008). As the core component of job burnout, emotional exhaustion can effectively predict job withdrawal and counterproductive behaviors, reducing individual and organizational performance and even endangering the organization’s future development (Maslach et al., 2001). As project managers or chief engineers, registered construction engineers are responsible for all aspects of a project, including cost, progress, quality, and risk management, leading teams ranging from dozens to thousands of people, managing funds from millions to billions of yuan, coordinating relationships among various project stakeholders, and implementing decisions and requirements from company management. Under such high job demands, insufficient job resources can easily lead to emotional exhaustion (Demerouti et al., 2001).

The “buffer” hypothesis of the JD-R model suggests that job resources can buffer the negative impact of high job demands on employees, mitigating the adverse effects of job demands (Bakker and Demerouti, 2007). Based on the source of resources, the resources individuals can draw upon during work are divided into external resources and internal resources. External resources mainly refer to social and organizational resources, such as job autonomy, high-quality superior-subordinate relationships, supervisor and coworker support, timely feedback, and career prospects; internal resources mainly refer to physiological and psychological resources, such as health, self-efficacy, self-esteem, and psychological capital (Hobfoll, 1989). Research by Xanthopoulou et al. (2007) shows that job resources moderate the impact of job demands on job burnout, weakening the negative effects of job demands under high resource conditions and enhancing them under low resource conditions. Yang et al. (2017) point out that the job demands (stress) of construction engineers mainly stem from the industry-specific, organizational, and social characteristics of their profession. The construction industry is highly regulated, with standardized procedures for most work processes and uniform, stringent, mandatory standards for product quality. Therefore, the job demands brought by industry characteristics are highly similar for every construction engineer, while organizational and social characteristics become the main factors differentiating individual work experiences. In organizations, leaders are typical representatives of the organization, and their style and behavior are important components of organizational characteristics, profoundly affecting their subordinates' work behavior and experiences. As the main manifestation of social characteristics in organizations, superior-subordinate and colleague relationships also significantly impact individual work experiences. Research by Bakker et al. (2005) shows that when employees have high-quality superior-subordinate relationships, high autonomy, timely feedback, and social support, negative factors such as work overload and emotional demands do not lead to high levels of burnout. Additionally, Xanthopoulou et al. (2007) point out that leadership support can provide a stronger buffer than coworker support, serving as a crucial resource in employees' work. However, other studies indicate that excessive attention and intervention from leaders can undermine employees' job autonomy, leading to burnout states such as emotional exhaustion and low achievement (Fernet et al., 2013). Based on this, this study constructs a moderated mediation model, using LMX as the independent variable, active management by exception (AMbE) as the mediating variable, and self-esteem as a relatively stable personality trait as the moderating variable, focusing on the impact of resource factors on construction engineers' work experiences (emotional exhaustion). This study makes three distinct contributions to the literature. First, while previous research has primarily examined leadership in construction settings through the lens of safety performance or project outcomes, we uniquely investigate its role in emotional exhaustion. This focus is particularly timely given the increasing recognition of mental health challenges in the construction industry. Second, we introduce a novel theoretical mechanism by examining how AMbE mediates the relationship between LMX and emotional exhaustion. This advances our understanding beyond simple direct effects, revealing the complex interplay between relationship quality and monitoring behaviors in construction management. Third, we challenge conventional wisdom regarding self-esteem’s protective role by demonstrating its amplifying effect on the relationship between monitoring behaviors and emotional exhaustion. This finding provides a new perspective on how personal resources function in high-pressure construction environments.

The Job Demands-Resources (JD-R) model provides a comprehensive theoretical framework for understanding workplace well-being and stress. According to this model, job characteristics can be classified into two categories: job demands and job resources. Job demands refer to physical, psychological, social, or organizational aspects of work that require sustained effort and are associated with certain physiological and psychological costs. In the construction industry, these demands include high workload, time pressure, complex technical requirements, and responsibility for safety and quality control. Job resources, conversely, are aspects that help achieve work goals, reduce job demands, or stimulate personal growth. For registered construction engineers, these resources include leadership support (manifested through LMX), professional autonomy, and personal resources such as self-esteem. The JD-R model is particularly relevant to the construction industry context as construction projects inherently involve intense time pressure and complex coordination requirements, creating significant job demands, where leadership support and effective management practices serve as crucial resources in managing these demands. The model effectively explains how the interaction between these industry-specific demands and resources affects outcomes like emotional exhaustion. In this study, we examine how the buffer hypothesis of the JD-R model operates in the construction industry context. LMX represents a crucial job resource that may buffer against job demands, while AMbE represents a potential job demand that requires sustained attention and effort from construction engineers. Following the buffer hypothesis, high-quality LMX relationships should help mitigate the impact of demanding leadership practices like AMbE on emotional exhaustion. Furthermore, we consider self-esteem as a personal resource that may influence how effectively job resources (LMX) can buffer against job demands (AMbE). This theoretical framework guides our investigation of how leadership quality, management practices, and personal resources interact to influence emotional exhaustion among registered construction engineers.

Leader-Member Exchange (LMX) was first proposed by Graen et al. (1972), who suggested that leaders have different exchange relationships with different subordinates. Graen et al. (1972), Graen and Uhl-Bien (1995) further proposed that leaders adopt differentiated management styles for different subordinates, treating them differently in terms of time, energy, and resources, establishing varying degrees of closeness. This means that leaders may trust and care more for certain subordinates, engaging in interactions beyond work relationships. These subordinates enjoy more material rewards, greater job autonomy, and more promotion opportunities. In other words, different employees have different job resources. If these employees perform the same job with the same job demands, the amount of job resources they receive may lead to different levels of emotional exhaustion. Research by Akbar and Akhtar (2018) indicates a significant negative correlation between LMX and emotional exhaustion. Research by Cheng et al. (2020) also shows that LMX significantly negatively predicts emotional exhaustion. Studies in the construction industry have highlighted the crucial role of leadership in managing work-related stress. Soeprapto demonstrated that LMX plays an important role in construction projects (Soeprapto, 2020). Yang et al. (2017) found that leadership support is a critical factor affecting job burnout among Chinese construction project managers. Based on this, we propose Hypothesis H1: LMX negatively predicts construction engineers' emotional exhaustion.

Management by exception originates from Taylor’s (F.W. Taylor) management thought, referring to a management system or principle where the management level handles routine work by providing handling suggestions for standardization (standardization, proceduralization) and then authorizes lower management personnel to handle it, while retaining the power to supervise the work of subordinates and focusing on exceptional work that cannot be standardized (Taylor, 1911). To ensure leaders have the time and energy to fulfill their leadership duties, the “exception principle” should be implemented, allowing enterprise leaders to reduce the need for daily repetitive work commands, concentrate on major issues, and enhance the independent work capability and responsibility of subordinates. Leaders can let departments handle institutionalized, proceduralized issues without intervention. Bass (1985) introduced the concept of management by exception into leadership research, defining it as a leadership style where leaders monitor employees' work processes to prevent errors and correct deviations and mistakes in a timely manner to ensure efficient task completion. Avolio and Bass (1998) further distinguished between active and passive management by exception. AMbE involves leaders actively observing employees' behaviors and correcting mistakes and problems in a timely manner to ensure effective task completion. Passive management by exception, on the other hand, involves leaders not intervening in subordinates' work as long as the current work situation is acceptable. AMbE involves leaders who proactively monitor followers' performance, anticipate problems, and take corrective actions before issues become serious. These leaders actively search for deviations from rules and standards, regularly track mistakes, and maintain vigilant supervision. In contrast, passive management by exception involves leaders who wait for problems to arise before taking action, intervening only when standards are not met or when mistakes have already occurred. Our study focuses specifically on AMbE because its proactive nature aligns with the high-stakes environment of construction projects, where anticipating and preventing problems is crucial for project success and safety. Furthermore, the construction industry’s emphasis on quality control, safety standards, and regulatory compliance makes AMbE particularly relevant as it involves systematic monitoring and quick correction of deviations from standards. Research by Stordeur et al. (2001) shows a significant positive correlation between AMbE and employees' emotional exhaustion. For registered construction engineers, there may also be a correlation between AMbE and emotional exhaustion.

Following the introduction of LMX by Graen et al., scholars worldwide have explained this phenomenon from different theoretical perspectives, summarized into three viewpoints: social exchange, role-making, and reciprocity continuum (Akbar and Akhtar, 2018). The social exchange perspective views the interaction between leaders and team members as a social exchange relationship. Based on exchange theory, social interactions involve economic and social exchanges. Economic exchanges are driven by interests, where people interact for mutual benefits, while social exchanges go further, based on mutual trust and extending beyond economic exchanges (Blau, 1964). Liden and Graen (1980) proposed that there are two types of exchange relationships based on their nature: formal relationships between leaders and subordinates within organizations, based solely on economic exchange; and social exchanges, where leaders and subordinates establish trust relationships for various reasons, involving emotional exchanges beyond formal relationships. Social exchanges transcend economic exchanges, forming closer relationships based on mutual trust and assistance. In high-quality leader-member relationships, leaders provide high levels of support and trust to subordinates. Due to these closer relationships and higher levels of trust, leaders may reduce active observation and monitoring of such employees. Within the construction industry context, research has revealed complex dynamics regarding leadership styles. Simmons et al. (2017) conducted a critical review of leadership paradigms in construction, highlighting how different leadership approaches, including transactional leadership behaviors like AMbE, influence project outcomes. Wu et al. (2016) further demonstrated how safety leadership works among different hierarchical levels in construction projects, emphasizing the importance of balanced leadership approaches in maintaining both performance and well-being. Based on this, we propose Hypothesis H2: AMbE mediates the relationship between LMX and emotional exhaustion.

Self-esteem, defined as an individual’s overall subjective evaluation of their own worth and the extent to which they view themselves as capable, significant, and worthy, plays an important moderating role in psychological and social life. It is an individual’s evaluation and perception of their own value, significantly influencing their behavior, emotions, and cognition (Rosenberg, 1965). First, self-esteem levels affect individuals' emotional experiences. High self-esteem individuals tend to have a more positive view of themselves and are more likely to experience self-esteem satisfaction and pleasure. Low self-esteem individuals, on the other hand, may feel more depressed, anxious, or inferior, generating negative self-evaluations (Brown and Marshall, 2006). Self-esteem helps individuals maintain emotional stability when facing setbacks and failures, reducing the impact of negative emotions (Baumeister et al., 2003). Therefore, self-esteem can mitigate the emotional impact of AMbE on construction engineers. Self-esteem also affects individuals' self-perception and self-evaluation. High self-esteem individuals usually have a higher evaluation of themselves, more likely to see their strengths and advantages. This positive self-perception helps them maintain confidence and better cope with life’s challenges. Low self-esteem individuals may focus more on their shortcomings and deficiencies, forming a negative self-perception that affects their self-confidence and self-esteem (Mruk, 2006). Therefore, high self-esteem individuals can maintain their confidence and self-esteem when facing AMbE, reducing the occurrence of emotional exhaustion. Additionally, self-esteem can stimulate individuals' intrinsic motivation, driving them to pursue higher goals and achievements. When individuals realize their value and abilities, they are more motivated to improve their skills and knowledge, achieving self-worth (Ryan and Deci, 2000). Self-esteem helps individuals adapt and integrate into social environments. High self-esteem individuals are generally more confident, open, and more likely to establish good interpersonal relationships. This good social adaptability helps them gain more support and resources in society (Leary and Baumeister, 2000). These factors help high self-esteem individuals adjust their emotional experiences when facing AMbE.

However, excessively high self-esteem may lead to problems such as arrogance, conceit, or overconfidence (Baumeister et al., 1996). These individuals may frequently feel the need for others' recognition and praise to maintain their high self-esteem. When not receiving enough affirmation, they may feel lost, uneasy, and even anxious (Crocker and Park, 2004). They may also be highly sensitive to external evaluations, overly concerned with others' opinions, interpreting normal, well-meaning comments or behaviors as threats to their self-esteem. This sensitivity can cause them to be overly tense in interpersonal interactions, potentially leading to conflicts (Kernis, 2003). Furthermore, since they often believe they are the best and most capable, they may experience severe self-doubt and depression when encountering failures or setbacks, struggling to accept their shortcomings, and even falling into self-denial (Neff and Vonk, 2009). Therefore, when facing excessive AMbE, high self-esteem individuals are more likely to experience emotional exhaustion. Therefore, we propose Hypothesis H3: Self-esteem moderates the impact of AMbE on emotional exhaustion.

In summary, this study focuses on Chinese registered construction engineers and proposes a moderated mediation model (as shown in Figure 1) to investigate the mediating role of AMbE in the relationship between LMX and emotional exhaustion and the moderating role of self-esteem.

This study selected Chinese registered construction engineers as the research subjects. We employed convenience sampling through professional networks and industry contacts in the construction sector. The participants were recruited from various construction companies and projects where we had established connections. This sampling approach, while not random, allowed us to access registered construction engineers who were actively engaged in project management roles. Before the survey, all participants were informed by the principal investigator about the purpose of the study, the voluntary nature of participation, the confidentiality of data handling, and their right to withdraw at any time. The expected time commitment for completing the survey was approximately 20–25 minutes. After obtaining informed consent, an anonymous survey was conducted between March and May 2023. A total of 400 questionnaires were distributed through both online and paper-based formats. We received 328 returned questionnaires, representing an initial return rate of 82%. After excluding 24 questionnaires due to incomplete responses or uniform answering patterns, 304 valid questionnaires were retained for analysis, yielding an effective response rate of 76%. The age distribution of the participants was as follows: 10 participants under 25 years old, 52 participants aged 26–35, 208 participants aged 36–45, and 34 participants over 45 years old. The educational background of the participants was: 98 with an associate degree or below, 178 with a bachelor’s degree, and 28 with a master’s or doctoral degree. The professional experience was distributed as follows: 52 participants with less than 2 years, 178 participants with 2–5 years, 50 participants with 6–10 years, and 24 participants with over 10 years.

The LMX scale compiled by Graen and Uhl-Bien was applied to measure LMX. The scale includes 7 items, such as “How well does your leader understand your work problems and needs?” and “How well does your leader recognize your potential?” All the items are scored on a 7-point Likert scale (ranging from 1 = not at all/None to 7 = Very High/Fully). Higher scores indicate a higher quality of the LMX relationship. We selected Graen and Uhl-Bien’s LMX-7 scale because it has been widely used across different cultural contexts and occupational groups, demonstrating good cross-cultural applicability. In the present study, the scale showed good internal consistency reliability (α = 0.916).

The AMbE dimension of the MLQ scale compiled by Bass was used, with subordinates evaluating the items(Sample items: “My supervisor focuses attention on irregularities, mistakes, exceptions, and deviations from standards” and “My supervisor keeps track of all mistakes”). The scale uses a 5-point Likert scale, with “1” indicating “strongly disagree” and “5” indicating “strongly agree.” Higher scores represent a higher degree of AMbE from the leaders. The AMbE dimension from the MLQ scale was chosen because it specifically measures monitoring and corrective behaviors, which are crucial for ensuring safety and quality standards in construction project management. In this study, the scale’s Cronbach’s alpha coefficient was 0.875.

The Rosenberg Self-Esteem Scale was used. This scale includes 10 items, such as “Overall, I am satisfied with myself.” It uses a 4-point Likert scale, with “1” indicating “strongly disagree” and “4” indicating “strongly agree.” Items 3, 5, 9, and 10 are reverse scored. Higher scores indicate higher levels of self-esteem. The Rosenberg Self-Esteem Scale was selected due to its stable psychometric properties and extensive validation in the Chinese context. In the present study, the scale’s Cronbach’s alpha coefficient was 0.843.

The Emotional Exhaustion subscale from the job burnout scale compiled by Li Yongxin was used. This subscale includes 5 items, such as “I often feel exhausted.” It uses a 7-point Likert scale, with “1” indicating “completely disagree” and “7” indicating “completely agree.” Higher scores represent higher levels of emotional exhaustion. We chose Li’s Emotional Exhaustion scale because it was specifically developed and validated for Chinese working populations. In this study, the scale’s Cronbach’s alpha coefficient was 0.875.

The data were analyzed using SPSS 25.0, Process v3.3 plugin, and bias-corrected Bootstrap techniques. Confirmatory factor analysis and homogeneity reliability analysis were used to verify the validity and reliability of the questionnaires. Pearson correlation analysis was used to explore the relationships between variables. Regression analysis and Bootstrap techniques were used to test the mediating role of AMbE between LMX and emotional exhaustion, the moderating role of self-esteem between AMbE and emotional exhaustion, and the moderated mediation effect of self-esteem on the relationship between LMX, AMbE, and emotional exhaustion.

The Harman single-factor test was used to detect common method bias (Podsakoff et al., 2003). The basic assumption of this technique is that if a significant common method bias exists, a single factor will emerge, or one general factor will explain the majority of the variance among the variables (Harman, 1976). An exploratory factor analysis was conducted with all variables in the hypothesized model, and the result showed five factors. The first factor explained 27.54% of the variance, which is less than 30%, indicating that there is no serious common method bias in the data.

The descriptive statistics and correlation matrix of the research variables are shown in Table 1. The results indicate that LMX is significantly negatively correlated with AMbE (r = −0.18, p < 0.01), Self-Esteem (r = −0.28, p < 0.01), and emotional exhaustion (r = −0.26, p < 0.01). AMbE is significantly positively correlated with Self-Esteem (r = 0.23, p < 0.01) and emotional exhaustion (r = 0.38, p < 0.01). Self-Esteem is significantly positively correlated with emotional exhaustion (r = 0.20, p < 0.01). The correlation analysis revealed several significant relationships between study variables. LMX showed a moderate negative correlation with emotional exhaustion (r = −0.26, p < 0.01), suggesting that as the quality of leader-member relationships increases, employees tend to experience less emotional exhaustion. The strength of this relationship indicates that while LMX is important, it explains approximately 6.8% of the variance in emotional exhaustion. LMX also demonstrated a negative correlation with AMbE (r = −0.18, p < 0.01), though the relationship was relatively weak. This suggests that leaders who develop higher quality exchanges with subordinates tend to rely slightly less on active monitoring and correction behaviors. However, the modest correlation indicates these leadership approaches are relatively independent. AMbE showed the strongest correlation with emotional exhaustion (r = 0.38, p < 0.01), explaining approximately 14.4% of the variance. This moderate positive relationship indicates that higher levels of monitoring and correction behaviors are associated with increased emotional exhaustion among construction engineers.

First, the independent variable (X), dependent variable (Y), mediating variable (W), and moderating variable (U) were standardized into Z scores (the variable names remained unchanged). Then, the Z scores were multiplied to generate the interaction term UW, and demographic variables such as age, marital status, and professional title were controlled for in the analysis. The hypotheses proposed in the preceding text could be examined through the following three equations and the results are shown in Table 2.

In Equation (1), LMX had a significant negative predictive effect on emotional exhaustion (β = −0.26, t = −4.62, p < 0.001), indicating that LMX inhibits emotional exhaustion. In Equation (2), LMX had a significant negative predictive effect on AMbE (β = −0.17, t = −3.21, p < 0.01), indicating that higher levels of LMX reduce the extent of AMbE received by construction engineers. In Equation (3), AMbE had a significant positive predictive effect on emotional exhaustion (β₂ = 0.45, t = 6.40, p < 0.001), suggesting that higher frequencies of AMbE are associated with higher levels of perceived emotional exhaustion. Therefore, AMbE mediates the relationship between job insecurity and emotional exhaustion. In Equation (3), the interaction term of AMbE and self-esteem had a significant positive predictive effect on emotional exhaustion (β₄ = 0.54, t = 3.26, p < 0.01), indicating that self-esteem moderates the relationship between AMbE and emotional exhaustion. Using model 14 in PROCESS v3.3 with bias-corrected non-parametric percentile Bootstrap, the results are shown in Table 3. When self-esteem is one standard deviation above the mean, AMbE mediates the relationship between LMX and emotional exhaustion, moderated by self-esteem (indirect effect = −0.12, accounting for 35.3% of the total effect). When self-esteem is at the mean level, the mediating effect of AMbE on the relationship between LMX and emotional exhaustion is moderated by self-esteem (indirect effect = −0.08, accounting for 26.7% of the total effect). When self-esteem is one standard deviation below the mean, the mediating effect of AMbE on the relationship between LMX and emotional exhaustion is not significant. This indicates that as self-esteem decreases, the mediating effect of AMbE on the relationship between LMX and emotional exhaustion gradually diminishes. Overall, self-esteem moderates the latter half of the mediation process (LMX → AMbE → emotional exhaustion), validating the proposed moderated mediation model.

To visually display the moderating effect of self-esteem, the impact of AMbE on construction engineers' emotional exhaustion under high and low self-esteem conditions was analyzed. Self-esteem was divided into high (M+1SD) and low (M−1SD) groups, and a simple slope test was conducted, producing a moderating effect analysis graph (Figure 2). As shown in Figure 2, regardless of high or low self-esteem levels, emotional exhaustion among construction engineers increases with higher levels of AMbE. Therefore, the direction of the impact of AMbE on emotional exhaustion is not altered by self-esteem levels. Under high self-esteem conditions, AMbE has a significant predictive effect on emotional exhaustion (β = 0.67, t = 6.69, p < 0.001), indicating that for construction engineers with high self-esteem, emotional exhaustion significantly increases with higher levels of AMbE. Under low self-esteem conditions, the predictive effect of AMbE on emotional exhaustion is less significant (β = 0.22, t = 2.25, p < 0.05), indicating that low self-esteem weakens the relationship between AMbE and emotional exhaustion. To further verify and visually display the moderated Mediating effect, the Johnson-Neyman method was introduced. The result shows that when the moderator value is below 1.41, the 95% CI includes zero, which means the mediating effect is not significant. A moderated mediation effect diagram using the Johnson-Neyman method was drawn (Figure 3). Therefore, self-esteem moderates the impact of AMbE on emotional exhaustion.

The path analysis revealed several key relationships. First, LMX demonstrated a significant direct negative effect on emotional exhaustion (β = −0.26, t = −4.62, p < 0.001). This substantial coefficient suggests that a one standard deviation increase in LMX quality is associated with a 0.26 standard deviation decrease in emotional exhaustion, holding other variables constant. In examining the indirect path through AMbE, we found that LMX negatively predicted AMbE (β = −0.17, t = −3.21, p < 0.01), indicating that higher quality relationships are associated with less monitoring behavior. Furthermore, AMbE positively predicted emotional exhaustion (β = 0.45, t = 6.40, p < 0.001), representing a stronger effect than the direct LMX-exhaustion relationship. The moderating effect of self-esteem proved substantial (β = 0.54, t = 3.26, p < 0.01). The positive coefficient indicates that self-esteem amplifies rather than buffers the relationship between AMbE and emotional exhaustion. Under high self-esteem conditions (+1 SD), the AMbE-exhaustion relationship is strong (β = 0.67, t = 6.69, p < 0.001). In contrast, under low self-esteem conditions (−1 SD), the relationship weakens considerably but remains significant (β = 0.22, t = 2.25, p < 0.05). This pattern suggests that construction engineers with higher self-esteem are more sensitive to monitoring behaviors, perhaps due to perceived threats to their professional autonomy. The indirect effect varies meaningfully across self-esteem levels, accounting for 35.3% of the total effect at high self-esteem levels and 26.7% at mean self-esteem levels, while becoming non-significant at low self-esteem levels. These findings reveal a complex interplay where the protective effects of LMX operate both directly and through reduced monitoring behaviors, but the effectiveness of these pathways depends significantly on the engineer’s level of self-esteem.

This study aimed to explore the impact of LMX on emotional exhaustion among Chinese registered construction engineers, focusing on the mediating role of AMbE and the moderating role of self-esteem. The primary findings can be summarized as follows:

Negative Correlation between LMX and Emotional Exhaustion: The study confirmed a significant negative relationship between LMX and emotional exhaustion, indicating that higher-quality exchanges between leaders and members reduce emotional exhaustion among construction engineers. This suggests that when leaders and subordinates engage in high-quality exchanges characterized by mutual trust, respect, and obligation, the emotional strain experienced by subordinates is significantly alleviated. This finding aligns with previous studies which have emphasized the protective role of quality leader-member interactions in mitigating stress and burnout in high-demand professions (Thomas and Lankau, 2009; Breevaart et al., 2015).

Mediating Role of AMbE: AMbE was found to mediate the relationship between LMX and emotional exhaustion. Specifically, high-quality LMX relationships reduce the extent of AMbE, which in turn lowers emotional exhaustion (Zhao et al., 2010). This mediation suggests that one mechanism through which LMX influences emotional exhaustion is by reducing the occurrence of AMbE behaviors. When leaders engage in high-quality exchanges, they are less likely to micromanage and more likely to empower their subordinates, thereby reducing the emotional toll on these professionals (Breevaart and Bakker, 2018; Gottfredson and Aguinis, 2017). This finding aligns with recent research on the importance of leadership styles in mitigating workplace stress and burnout (Yang et al., 2017; Montano et al., 2017; Inceoglu et al., 2018).

Moderating Role of Self-Esteem: Analysis revealed that self-esteem moderated the relationship between AMbE and emotional exhaustion. High self-esteem exacerbates the negative impact of AMbE on emotional exhaustion, while low self-esteem buffers this impact (Orth and Robins, 2021). This finding highlights the complex role of individual differences in stress responses and suggests that self-esteem can either amplify or mitigate the effects of leadership behaviors on emotional well-being (Aronson et al., 2006). High self-esteem individuals, while typically resilient, may interpret AMbE as a threat to their autonomy and competence, thus experiencing higher emotional exhaustion under such conditions (Dust et al., 2018; Zacher and Rudolph, 2021). This aligns with recent research on the nuanced effects of self-esteem in organizational contexts (Zheng et al., 2015).

Our findings advance the theoretical understanding of emotional exhaustion in construction settings in three significant ways. First, we extend the JD-R model (Bakker and Demerouti, 2007) by demonstrating how leadership behaviors can function as both resources and demands within the same organizational context. Unlike previous studies that primarily treated leadership support as a resource (Bakker et al., 2005; Fernet et al., 2013), our findings reveal that leadership behaviors can simultaneously enhance and deplete employees' psychological resources, depending on their manifestation as either LMX or AMbE. Second, our research provides new insights into the psychological mechanisms through which leadership affects emotional exhaustion in highly regulated industries. While Yang et al. (2017) identified work stress as a critical challenge in construction settings, our study demonstrates how leadership practices mediate this relationship. This advances theoretical understanding beyond simple direct effects to reveal the complex pathways through which leadership influences emotional exhaustion. Third, our findings challenge the traditional understanding of personal resources in the JD-R model. Previous research following Hobfoll’s (Xanthopoulou et al., 2007) conservation of resources theory typically viewed self-esteem as a protective factor. However, our results suggest that in the unique context of construction engineering, high self-esteem can amplify rather than buffer the negative effects of certain leadership behaviors. This finding extends theoretical understanding of how personal resources function in highly professional contexts where autonomy and expertise are particularly valued.

This study extends the Job Demands-Resources (JD-R) model by incorporating LMX and AMbE as critical factors influencing emotional exhaustion (Bakker and Demerouti, 2017). The findings suggest that leadership quality (LMX) and management style (AMbE) are significant resources and demands, respectively, that affect the emotional well-being of construction engineers. High-quality LMX serves as a valuable resource that can mitigate job demands and reduce emotional exhaustion (Zhang et al., 2014). By integrating these elements into the JD-R framework, this study provides a more nuanced understanding of how leadership dynamics function as both resources and demands in high-stress work environments (Schaufeli, 2017). This extension of the JD-R model underscores the importance of considering interpersonal relationships and management practices when assessing stress and burnout in professional settings (Lesener et al., 2019; Taris et al., 2017).

The study highlights the dual role of leadership in influencing employee well-being. While positive LMX relationships provide emotional support and reduce exhaustion, (Breevaart and Bakker, 2018) AMbE can have the opposite effect, particularly for individuals with high self-esteem (Zheng et al., 2015). This duality underscores the complexity of leadership dynamics and the need for balanced management approaches that consider both supportive and corrective behaviors (Hoch et al., 2018). Leaders must navigate the fine line between providing necessary oversight and fostering an environment of trust and autonomy (Gottfredson and Aguinis, 2017). This dual role suggests that leadership effectiveness is contingent upon the ability to tailor management practices to the individual needs and characteristics of employees, thereby maximizing support while minimizing unnecessary stress (Inceoglu et al., 2018; Montano et al., 2017).

The moderating role of self-esteem adds a nuanced understanding of individual differences in stress responses. High self-esteem, typically considered a protective factor, can amplify stress under certain management styles (e.g. AMbE). This finding challenges the simplistic view of self-esteem as universally beneficial and suggests that its impact is context-dependent (Orth and Robins, 2021). For instance, individuals with high self-esteem may thrive in environments that recognize and utilize their strengths but may struggle when their autonomy is perceived to be undermined (Dust et al., 2018). Conversely, those with lower self-esteem may benefit from structured guidance and clear expectations provided through AMbE (Zacher and Rudolph, 2021). This complexity indicates that self-esteem influences not only how individuals perceive and react to job demands but also how they leverage available resources, thus affecting their overall emotional resilience (Arslan et al., 2021).

Organizations should invest in training programs that enhance the quality of LMX. Leaders should be trained to build trust, provide support, and recognize the contributions of their subordinates. High-quality LMX relationships can create a more supportive work environment, reducing emotional exhaustion and improving overall job satisfaction. By fostering environments where leaders and subordinates engage in meaningful, trust-based exchanges, organizations can enhance employee well-being and productivity. Practical steps include leadership development programs that emphasize emotional intelligence, conflict resolution, and communication skills, all of which are essential for nurturing high-quality LMX relationships.

While AMbE is necessary for maintaining high standards and correcting errors, it should be balanced with supportive behaviors. Leaders should be cautious not to overuse corrective actions, especially with employees who have high self-esteem, as this can lead to increased emotional exhaustion. A balanced approach that includes both supportive and corrective elements is essential for maintaining employee well-being. Leaders should be trained to recognize the signs of burnout and to adjust their management style accordingly. Implementing feedback mechanisms where employees can voice their concerns and preferences regarding management practices can also help in striking the right balance.

Interventions aimed at reducing emotional exhaustion should consider individual differences in self-esteem. For employees with high self-esteem, it may be beneficial to provide autonomy and opportunities for self-directed work, thereby reducing the need for AMbE. For those with low self-esteem, supportive interventions that build confidence and provide positive feedback can be more effective. Tailoring interventions to the self-esteem levels of employees can enhance their effectiveness and ensure that the unique needs of each individual are met. This approach requires a nuanced understanding of the workforce and may involve personalized development plans, mentoring programs, and regular assessments of employee well-being and engagement.

While our sample consisted exclusively of registered construction engineers, the theoretical mechanisms studied (LMX, AMbE, self-esteem, and emotional exhaustion) are fundamental organizational and psychological processes. The relationships between these variables likely exist in other professional contexts, though their strengths may vary. However, the specific work conditions, responsibilities, and pressures faced by registered construction engineers may influence these relationships in unique ways, such as through the heightened importance of leadership support in high-risk construction environments.

While our sample size (N = 304) was adequate for statistical analyses, a larger sample size would provide greater statistical power and enable more sophisticated analyses. The cross-sectional design of this study limits the ability to make causal inferences. Longitudinal studies are needed to establish causal relationships between LMX, AMbE, self-esteem, and emotional exhaustion. Future research could also explore potential reciprocal relationships, where emotional exhaustion impacts LMX and management behaviors over time. Longitudinal designs would enable researchers to track changes in these variables over time, providing deeper insights into the temporal dynamics and causal pathways involved. Additionally, experimental or quasi-experimental designs could be employed to test specific interventions aimed at improving LMX and reducing emotional exhaustion. Meanwhile, this study focused on Chinese registered construction engineers, a specific professional group with unique job demands and cultural contexts. Although registered construction engineers represent a distinct professional group, they play a crucial role in China’s construction industry development. Since 2008, Chinese regulations have mandated that construction project leaders must be registered construction engineers, highlighting their essential management responsibilities and technical expertise. The mechanisms we found regarding relationships between LMX, AMbE, and emotional exhaustion need to be validated among a broader range of construction industry professionals to determine their generalizability. Future research should examine whether the findings can be generalized to other professions and cultural settings. Additionally, exploring other contextual factors, such as organizational culture and industry-specific demands, could provide a more comprehensive understanding of the dynamics at play. For instance, different cultural norms regarding leadership and management practices may influence the effectiveness of LMX and AMbE. Comparative studies across different countries and industries could illuminate these contextual influences and help tailor interventions to specific settings.

This study sheds light on the complex interplay between leadership quality, management style, and individual differences in influencing emotional exhaustion among Chinese registered construction engineers. By integrating the JD-R model with LMX theory and considering the moderating role of self-esteem, the study provides valuable insights for both theory and practice. Enhancing LMX quality, balancing management styles, and tailoring interventions based on self-esteem are critical strategies for reducing emotional exhaustion and promoting the well-being of construction engineers. Future research should continue to explore these dynamics in different contexts and with larger samples to further validate and extend these findings. By doing so, we can develop more effective strategies for managing stress and enhancing the well-being and performance of professionals in high-stress industries. This comprehensive approach will not only improve individual health and job satisfaction but also contribute to the overall sustainability and productivity of organizations.