Next Article in Journal
Screen Façade Pattern Design Driven by Generative Adversarial Networks and Machine Learning Classification for the Evaluation of a Daylight Environment
Previous Article in Journal
A State-of-the-Art Review on the Application of Geopolymerization/Geopolymer in Environmental Fields
Previous Article in Special Issue
Integrating Panic Behavior into Agent-Based Subway Evacuation Simulations
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Buildings
Volume 15
Issue 22
10.3390/buildings15224055
buildings-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Causal Model Analysis of the Impact of Formalism, Psychological Contract and Safety Coaching on Safety Compliance and Participation in Taiwan

by
Chi-Jan Huang
1,
Tsung-Lin Wu
2 and
Hsiang-Te Liu
3,*
1
General Education Center, Ming Chuan University, Taipei 111, Taiwan
2
Program in International Business Administration, I-Shou University, Kaohsiung 840, Taiwan
3
Department of Public Affairs and Administration, Ming Chuan University, Taoyuan 333, Taiwan
*
Author to whom correspondence should be addressed.
Buildings 2025, 15(22), 4055; https://doi.org/10.3390/buildings15224055
Submission received: 16 September 2025 / Revised: 2 November 2025 / Accepted: 8 November 2025 / Published: 10 November 2025
(This article belongs to the Special Issue Human Factor on Construction Safety)
Browse Figures
Versions Notes

Abstract

This study primarily investigates the influence of safety coaching, psychological contract, and safety knowledge on safety compliance and safety participation among construction workers. Specifically, the research examines the effect of safety formalism on workers’ psychological contract, which in turn influences their safety behaviors on-site. The issue of construction workers not strictly adhering to safety regulations is a topic that has rarely been explored by past researchers of construction safety behavior. To address this, the study collected 99 valid samples and utilized confirmatory factor analysis (CFA) to confirm the reliability and validity of the questionnaire. A structural equation model (SEM) was then employed to test the established hypotheses. The research findings confirm, as follows: Safety formalism negatively influences psychological contract. Safety coaching positively influences safety knowledge. Safety knowledge and resilience positively influence both safety compliance and safety participation. Safety compliance positively influences safety participation.

1. Introduction

“Safety behavior” refers to behavior that reduces unsafe incidents, accidents, and injuries [1,2]. On construction sites, safety behavior ensures and enhances workplace safety [3]. Safety behavior comprises safety compliance and safety participation [4].
Safety compliance is regarded as an attitude of adhering to safety procedures and is considered an in-role behavior. Safety participation involves assisting colleagues and improving workplace safety, which is considered an extra-role behavior [5]. Safety compliance includes adherence to mandatory regulations and standard operating procedures. On construction sites, safety compliance entails following work policies, wearing protective equipment, and heeding organizational safety requirements. Safety participation, which is more voluntary in nature, involves participation in safety-related activities, such as offering safety suggestions, encouraging safety learning and training, and prioritizing safety issues [6].
Safety participation on construction sites may include attending meetings, assisting coworkers with safety tasks, and proactively offering safety suggestions [7]. Safety participation does not fully fall within the scope of official workplace policies; rather, it closely resembles organizational citizenship behavior [8]. Indeed, some studies also refer to safety participation as “safety citizenship behavior” [9]. From the perspective of social identity theory, an individual’s self-concept is formed from the knowledge of their membership in social groups. These social forces motivate construction workers to voluntarily engage in safety-related activities. The construction industry is characterized by a high level of occupational hazards. Both safety compliance and safety participation have been empirically linked to reductions in workplace accidents [10]. Compliance with safety procedures has been negatively associated with construction site accidents [11], and safety participation contributes to reduced safety incidents in construction contexts [12].
The construction site safety explored in this study is also related to safety climate theory and high-reliability organization theory. High-reliability organizations also emphasize the need for a reluctance to simplify work procedures [13]. Formalism in safety regulations tends to simplify, contradicting the goals of high-reliability organizations. Furthermore, high-reliability organizations emphasize commitment to resilience, requiring workers to continuously review, detect, and correct safety issues as they arise [13]. This also increases worker participation and compliance in high-reliability organizations. Safety climate is a subcategory of general organizational climate. It is a shared understanding that prioritizes compliance with safety policies, regulations, and procedures [14]. While safety climate can be a factor in resistance to change, high-reliability organizations emphasize continually challenging existing operations to achieve long-term stability and safety [14]. High-reliability organizations emphasize communication between managers and workers, and worker participation in planning safe operations is a crucial factor in ensuring site safety.
Psychological capital is viewed as a positive capacity developed through the process of personal growth and development [15]. Its components include self-efficacy and resilience [16]. Psychological capital also serves as a positive force driving employee performance and job satisfaction [17]. Moreover, it has been found to enhance organizational citizenship behavior [18]. The influence of psychological capital on safety performance may vary depending on contextual risks and specific tasks [10]. In the present study, resilience was adopted as a key dimension to examine its impact on safety compliance and safety participation.
According to the conservation of resources theory, individuals strive to retain valuable resources [19]. Possessing greater psychological resources such as tolerance and stress resistance enables individuals to develop stronger resilience. Individuals with higher resilience are better able to adapt to hazardous and high-pressure work environments [20]. Construction workers, likewise, preserve valuable resources to strengthen their resilience.
“Resilience” refers to an individual’s ability to maintain strong determination when facing unexpected and unforeseen situations [15]. Those with higher resilience not only learn from failure and accept criticism but also adapt quickly to changing environments [21,22]. On construction sites, unexpected and unforeseen situations occur frequently. The influence of resilience on employee behavior has been widely examined [23], and the role of construction workers’ resilience in shaping safety compliance and safety participation has also been investigated [24,25]. A key characteristic of a High-Reliability Organization (HRO) is its commitment to resilience [13]. Employees exhibiting high levels of resilience are generally more willing to participate in and comply with safety regulations.
Previous research has demonstrated that safety knowledge is an antecedent of organizational safety behavior [10]. On construction sites, even small errors may result in serious injuries or fatalities. “Safety knowledge” refers to the extent of an individual’s understanding of safety regulations and standard operating procedures [26]. Christian et al. also indicated that occupational safety knowledge serves as an antecedent of both safety compliance and safety participation [10]. Vinodkumar and Bhasi further identified safety knowledge as a critical factor influencing safety [27]. Causes of construction accidents have been found to include safety participation, safety knowledge, and compliance with safety regulations [23].
A distinctive feature of the present study is the incorporation of formalism in safety regulations and policies into the investigation of construction site safety behavior. “Formalism” refers to the discrepancy between safety regulations or policies and actual safety behaviors. Previous studies on formalism have been limited, and most were restricted to qualitative exploration. Milne noted that Riggs’ administrative ecology model for developing countries lacked empirical validation to evolve into a middle-range theory [28]. Other scholars have also emphasized that formalism, when examined from the perspective of administrative ecology, has been insufficiently supported by empirical evidence [29]. The present study therefore attempts to explore the role of formalism in construction site safety behavior from an empirical standpoint. According to Milne, Riggs’ administrative model for developing countries comprises formalism, ritualistic methods, and centralization [28]. Formalism specifically refers to the gap between regulatory provisions and their actual implementation [28]. Riggs further argued that in developing countries, weak enforcement pressure on policy goals and the lack of supervision from social forces result in laxity in policy implementation and management [30]. Consequently, Riggs asserted that policies and regulations alone fail to bring about substantive behavioral change [30]. Additionally, in countries characterized by formalism, superiors often refrain from fully delegating authority to subordinates, which results in a lack of shared values between implementers and managers [28]. Based on his observations, Riggs concluded that public policies and regulations in developing countries are often ineffective in practice [28]. The formalism of construction site safety policies tends to result in lax enforcement, which in turn diminishes the reciprocity norms between the organization and site workers. In particular, when there is a lack of shared understanding and consistency between management and construction workers regarding safety policies and measures, the psychological safety contract between workers and the organization is undermined.
Although Taiwan has gradually modernized and democratized, the phenomenon of formalism in public administration continues to persist. One example is the persistence of rating errors in the performance appraisal system—a form of formalism. In practice, Taiwanese civil servants often rely on the so-called “unwritten rule” of alternately assigning A and B ratings rather than basing appraisals on actual job performance. Moreover, the proportion of A-level ratings is capped at 75%, meaning that nearly 75% of civil servants receive the highest performance grade A [31]. As a result, performance evaluations and promotions among civil servants have become largely perfunctory, with appraisals and promotions losing their intended incentive effects. A central question is why supervisors of civil servants across Taiwan have been unable to break away from such irrational performance appraisal practices. According to Tseng and So, entrenched organizational politics and cultural habits have led to a disjunction between the legal framework of the appraisal system and its actual operation. Although Taiwan’s political and administrative development have diverged significantly from when it was regarded as a developing country 60 years ago, this form of appraisal formalism among civil servants persists, primarily because, as Tseng and So observed, organizational politics and organizational culture remain difficult to change [31]. When administrative agencies fail to implement and enforce safety regulations and inspections, it fosters a perception among construction workers that safety regulations are merely perfunctory. Scholars have previously noted that an organizational climate’s resistance to change can persist for a substantial duration [14]. When a construction organization’s safety plan is consistently poorly executed over the long term, it negatively impacts the psychological contract felt by site workers.

2. Literature Review and Hypotheses Development

2.1. The Relationship Between Formalism and the Safety Psychological Contract

Bureaucratic systems characterized by high levels of formalism have been associated with a lack of dedication among public servants [32]. In developing countries, public servants are often perceived as passive and non-participatory, leaving decisions to senior administrators [32]. They tend to adhere rigidly to outdated procedures without adopting innovative practices or embracing new ways of thinking. Another reason public servants refrain from problem solving is that supervisors are reluctant to delegate authority to subordinates. Consequently, civil servants may be unwilling to assume responsibility, placing personal interests above public interests [32]. Similarly, when construction workers perceive safety policies as impractical, their sense of a safety psychological contract naturally diminishes. They may feel that safety regulations implemented by the organization are insufficient to ensure genuine safety.
Shamsul Haque argued that public servants in developing countries often lack professional skills, and their practices are highly formalized [29]. Rodman pointed out that inconsistencies between policy formulation and implementing agencies frequently result in poor coordination [33]. Moreover, excessive paperwork and bureaucratic red tape often overshadow regulatory and policy objectives [33]. The prevalence of formalism therefore prevents effective policy implementation [33]. In the context of construction sites, when workers perceive safety policies as excessively formalistic, their safety psychological contract is naturally weakened. On construction sites, numerous occupational safety and health laws and building regulations must be observed, and personal protective equipment and safety facilities require attention. However, when workers perceive that regulatory and policy objectives are not genuinely prioritized, they conclude that governments and organizations have failed to fulfill their obligations for safe construction practices.
Hypothesis 1.
Safety formalism negatively influences the safety psychological contract.

2.2. The Relationship Between Safety Coaching and Safety Knowledge

“Safety coaching” is defined as the behavior of guiding individuals to apply analytical techniques and interpersonal interactions in order to create a safer environment [34]. Burke, Crowe, Salvador, and Chan-Serafin conducted a meta-analysis on safety training and safety knowledge, which revealed that engaging and interactive safety training enables individuals to better cope with high-risk situations [35]. Training approaches that involve modeling, simulation, and practice were found to be more effective forms of coaching.
“Behaviorally oriented safety coaching” refers to the process of providing support for safe behaviors and offering constructive feedback in response to risky workplace behaviors [36]. As Geller et al. noted, construction firms that place emphasis on hazard reduction and worker care tend to engage more actively in safety coaching [36]. Safety coaching enhances workers’ understanding of safety knowledge. According to Geller, promoting workers’ safety knowledge through safety coaching is crucial for workplace safety [37].
Hypothesis 2.
Safety coaching positively influences safety knowledge.

2.3. The Relationships Between Safety Knowledge, Safety Participation, and Safety Compliance

“Safety knowledge” refers to the extent of an individual’s understanding of safety procedures, training, and operating instructions [38]. Christian et al. identified safety knowledge as an antecedent of both safety compliance and safety participation [10]. Their meta-analysis reported a correlation coefficient of approximately 0.6 between safety knowledge and safety participation.
Greater knowledge of work processes and hazards reduces the likelihood of accidents on construction sites [39]. Prior studies have also indicated that safety knowledge enhances employees’ safety participation [10]. Griffin and Neal noted that safety knowledge is an antecedent of safety compliance, as workers must possess safety knowledge before they can comply with safety regulations [10]. When workers possess deeper safety knowledge, they are more willing to engage in safety compliance and safety participation.
Hypothesis 3.
Safety knowledge positively influences safety compliance.
Hypothesis 4.
Safety knowledge positively influences safety participation.

2.4. The Relationships Between Resilience, Safety Participation and Safety Compliance

“Resilience” can be defined from the perspectives of outcome, capability, and process. From an outcome perspective, resilience refers to an individual’s ability to adapt and thrive despite exposure to severe threats [40]. From a capability perspective, resilience is the capacity to rapidly adapt and innovate in the face of significant change [41,42]. From a process perspective, resilience is the adaptive and restorative process through which individuals respond to trauma, adversity, threats, or stressors [43].
Psychological resilience functions as a protective mechanism when facing stress [44,45,46]. Individuals with higher resilience are better equipped to cope with demanding work challenges and pressures [47,48]. Organizations can adopt structural controls and procedural controls at construction sites to reduce injuries and accidents. Structural controls correspond to safety compliance, involving long-term planning and design, whereas procedural controls are immediate safety requirements [49].
Resilience has been regarded as a pathway to safety improvement [50]. It is particularly relevant in highly complex workplace environments, especially in industries marked by interacting internal risk factors, high uncertainty, and volatility [51]. Construction sites themselves are complex, dynamic, and unstable environments, and workers require substantial resilience to cope with safety-related incidents [51]. Workers with stronger resilience are more inclined to engage in both safety compliance and safety participation.
Hypothesis 5.
Resilience positively influences safety compliance.
Hypothesis 6.
Resilience positively influences safety participation.

2.5. The Relationship Between Safety Psychological Contract and Safety Participation

According to the social exchange theory [52], psychological contracts represent reciprocal exchange relationships between individuals and organizations [53]. When employees perceive that organizations are fulfilling their obligations by providing satisfactory incentives, they develop positive psychological contracts [54]. Unlike formal written employment contracts, psychological contracts stem from verbal communication, observation, and organizational culture [55]. Psychological contracts play an important role not only in everyday work but also in complex and dynamic environments [56].
The safety psychological contract refers to an individual’s explicit or implicit commitment to safety obligations [57]. Such contracts are believed to influence construction workers’ safety participation and safety performance. Prior studies have found that the safety psychological contract is a significant predictor of safety performance [58]. When employees perceive that the organization has fulfilled its obligations, they are more likely to reciprocate with behaviors that benefit the organization [59]. From the perspective of social interactions between employees and organizations, organizational support for safety policies enhances employees’ safety participation [60]. When employees are satisfied with organizational safety policies, they are more willing to engage in and comply with safety-related activities [58].
Hypothesis 7.
The safety psychological contract positively influences safety participation.

2.6. The Relationship Between Safety Compliance and Safety Participation

Safety performance consists of safety compliance and safety participation [23]. “Safety compliance” refers to carrying out workplace safety tasks, such as using personal protective equipment and adhering to safety regulations. “Safety participation” involves contributing to workplace safety, e.g., attending routine safety meetings and writing safety reports. Multiple studies have identified safety compliance as an antecedent of safety participation [61,62,63].
When workers experience job insecurity or fear due to workplace uncertainty, they are less willing to attend safety training sessions and workshops [64]. In contrast, when construction workers pay greater attention to safety compliance, their safety participation behaviors increase. For example, workers who are willing to wear protective equipment are also more likely to attend safety workshops. Workers’ safety compliance has consistently been shown to promote safety participation [65]. In other words, when workers are willing to engage in safety compliance, their level of safety participation increases.
Hypothesis 8.
Safety compliance positively influences safety participation.
The relationships between all the research hypotheses are shown in Figure 1.

3. Materials and Methods

3.1. Sample, Tools, and Procedure

This study was conducted at construction sites in Taiwan. Utilizing a convenience sampling approach, a total of 250 questionnaires were distributed. From these, 99 valid responses were collected, resulting in a response rate of 39.6%. Using G*Power 3.1.9.7, the required sample size for this study was calculated to be 110, with setting α to 0.05 and Power (1-β err prob) at 0.95, effect size to 0.10, and the number of independent variables to 6. The sample size of 99 is slightly lower than the estimated required sample size of 110. The limited time, financial resources and willingness of construction workers to respond are the reasons why the sample size of this study is slightly low. A wave analysis was conducted to examine potential non-response bias. The sample was segmented into two waves based on the submission order: the first 49 responses and the latter 50 responses. Independent samples T-tests were performed across all study constructs. As the significance levels (p-values) ranged between 0.14 and 0.94, no statistically significant differences were observed between the early and late respondent groups. Therefore, non-response bias is considered minimal in this study.
For the gender composition of the sample, males accounted for 66.7% and females for 33.3%. In terms of seniority, respondents with 1–3 years of experience constituted the largest group at 39.4%, followed by those with 4–7 years of experience at 21.2%. Experienced employees with 16 years or more of tenure also made up a notable portion of the sample, at 18.2%. Regarding job titles, manual and technical roles were more prevalent. The highest percentage of respondents were laborers (25.3%), followed by engineers (18.2%) and technicians (15.2%) (as shown in Table 1). Other roles, such as architects, managers, and administrative staff, represented a smaller proportion of the sample. Due to the project-based and seasonal nature of the construction industry, along with challenging work environments, occupational safety risks, and reliance on foreign labor, workforce turnover at construction sites tends to be high. In the absence of annual statistical data from government agencies, this study was unable to conduct convenience sampling validation or population-level testing.
Prior to beginning the questionnaire, participants were provided with an informed consent statement. All participant data will be kept confidential in the research laboratory for a period of one year before being destroyed. This study is intended solely for academic journal submission, with no commercial interests involved. Participants are free to withdraw from the survey at any time without any pressure. In accordance with Article 6 of the researcher’s University Human Subject Research Ethics Committee Procedures, research concerning new medical technologies, new medicines, new medical devices, or the bioavailability and bioequivalence of generic drugs on human subjects, when based on medical theory, requires mandatory approval by the Institutional Review Board (IRB). As this study does not involve drugs or clinical human experimentation, and informed consent was obtained from all respondents, the ethical requirements were fulfilled.

3.2. Measures

Safety compliance was measured using items adapted from the scales by Griffin and Neal [23] and Vatankhah [66]. Example items include, as follows: “I perform my work tasks in a safe manner,” “I use safety protective equipment while working,” and “I follow safety procedures while working.” Safety participation was also assessed using items adapted from the scales by Griffin and Neal [23] and Vatankhah [66]. Example items from this scale are: “I proactively carry out tasks to improve site safety,” “I am willing to assist my colleagues when they are in danger on-site,” and “I am willing to put in extra effort to maintain site safety.” Safety knowledge was measured using items adapted from the scales by Griffin and Neal [23] and Vatankhah [66]. Example items include, as follows: “I know how to carry out work on the jobsite safely,” “I know how to use safety equipment and standard work procedures,” “I know how to maintain or improve workplace health and safety,” “I have the knowledge to reduce the risk of accidents in the workplace.” (Table A1).
Safety coaching was assessed using a modified version of the scale by Wu et al. [67]. Example items include, as follows: “The organization educates workers about the importance of site safety,” “The organization clearly explains site safety concepts,” and “The organization does not describe a vision for what safety looks like (R).” Safety psychological contract was measured using items adapted from Newaz et al. [68]. Example items are as follows: “The organization invites workers to participate in safety decision-making meetings,” “The organization listens to workers’ safety-related opinions,” “The organization conducts an investigation when a site safety accident occurs,” and “The organization arranges regular safety training.”
Resilience was measured using a modified version of the scale developed by Shin et al. [69]. Example items include, as follows: “I am able to cope with changes in the environment,” “No matter what happens, I am able to handle it,” and “I can still achieve my goals even when facing difficulties.” Safety formalism was operationalized based on Riggs’s [30] definition. The items were as follows: “I feel that safety regulations and their actual implementation are not exactly the same,” “I feel that safety laws on the construction site are sometimes difficult to fully enforce,” “I feel that many safety systems are not easy to implement,” and “I feel there are discrepancies between construction site safety regulations and their current implementation status.” The questionnaire in this study was designed with a total of four reverse-coded items. Specifically, one item was reverse-coded for the safety coaching construct, two items for safety compliance, and one item for safety participation.
All constructs in this study were measured using a 7-point Likert scale, with the exception of safety coaching, which used a 5-point Likert scale. The scales ranged from 1 (strongly disagree) to 7 (strongly agree). Cronbach’s α values for all constructs ranged from 0.87 to 0.95 (Table 2), which is higher than the recommended threshold of 0.6 suggested by previous statistics expert [70]. This study was grounded upon a well-established scale, and the questionnaire underwent stringent examination for both content validity and reliability. Beyond facilitating comparisons with previous research, this approach effectively allowed the investigation to explore novel theoretical relationships.

3.3. Detection of Common Method Variance (CMV)

In terms of research methodology, common method variance (CMV) is a form of measurement error. This internal inconsistency, caused by CMV, is a threat that needs to be controlled [70,71]. Self-report questionnaires are considered one of the sources of CMV. To mitigate this issue, the survey items in this study were designed to be simple, easy to understand, and easy to answer, thus reducing the CMV problems caused by respondent misinterpretation. Furthermore, this study employed anonymous responses and mixed 5-point and 7-point Likert scales, which are also believed to help lower CMV issues [71].
To assess common method variance (CMV), Harman’s single-factor test was performed [72]. Using unrotated exploratory factor analysis, the first factor accounted for 49% of the variance. Since the variance explained by the first factor did not exceed 50%, it indicates that CMV is not a serious concern in this study.
This study implemented the Unmeasured Marker Variable (UMV) approach by including a construct theoretically unrelated to the primary model. Model 1 was established by linking this marker variable to all main research constructs; the fit indices indicated excellent fit (CFI = 0.98; RMSEA = 0.017). Model 2 was then specified by constraining the relationship between the marker variable and all research constructs to zero, also demonstrating an excellent fit (CFI = 0.99; RMSEA = 0.012). The differences in factor loadings for all items between Model 1 and Model 2 ranged from 0 to 0.005. Given the minimal change in both model fit indices and factor loadings across the two models, it is concluded that the results of this study are not substantially contaminated by common method variance (CMV) bias.

3.4. Validity and Reliability Analysis

This study employed confirmatory factor analysis (CFA) to assess the reliability and validity of the research factors [73]. Regarding the parsimonious fit of the model, the parsimony normed fit index (PNFI) was 0.62, exceeding the recommended threshold of 0.50. In terms of overall model fit, the incremental fit index (IFI) was 0.99, the comparative fit index (CFI) was 0.99, the non-normed fit index (NNFI) was 0.99, and the goodness-of-fit index (GFI) was 0.99, all surpassing the required cutoff value of 0.90. Furthermore, the standardized root mean square residual (SRMR) of the research framework was 0.06, which is below the acceptable threshold of 0.08. The ratio of chi-square to degrees of freedom (χ2/df) was 1.02, and the 90% confidence interval (CI) for the Root Mean Square Error of Approximation (RMSEA) ranged from 0.00 to 0.04. Collectively, these fit indices confirm that the conceptual model demonstrates a good fit with the data.
The factor loadings λ in this study ranged from 0.74 to 0.94, which is higher than the commonly accepted minimum value of 0.5 in academic research [74]. The Z-values for all factor loadings were statistically significant, indicating that the research constructs possess both convergent and construct validity. All items across the research constructs were retained without any deletions. Composite reliability (CR), which assesses the internal consistency of latent factor items, ranged from 0.87 to 0.95. This is also above the widely accepted threshold of 0.7 [74], confirming the internal consistency of all latent variables in this study.
Average variance extracted (AVE) is a measure of the extent to which a latent factor explains the variance of its observed items. The AVE values for all latent factors in this study ranged from 0.63 to 0.84, exceeding the widely accepted minimum threshold of 0.5. Evidence suggests that the convergent and discriminant validity of all latent factors in this study.
The square root of the average variance extracted (AVE) is considered an indicator of a latent factor’s discriminant validity if it is higher than the correlation coefficients [75]. In Table 3, the correlation coefficients are located in the bottom-left corner, and the square roots of the AVEs are diagonally located. The square roots of the AVEs in this study ranged from 0.79 to 0.92, all of which were higher than the corresponding correlation coefficients. Additionally, the top-right corner of the table shows the heterotrait–monotrait (HTMT) ratio of correlations. The HTMT values for all latent factors in this study were below 0.90 (Table A2), which is considered evidence of discriminant validity [76]. Finally, the AVE values in this study were also greater than the maximum shared variance (MSV) and the average shared variance (ASV), further confirming the discriminant validity of the latent factors [70].
Because the correlation between safety coaching and safety knowledge reached 0.88 and the HTMT value between the two constructs was 0.879, which is below the 0.90 threshold, the study proceeded by comparing two models: an unconstrained (freely estimated) Model 1 and a constrained Model 2, in which the correlation between safety coaching and safety knowledge was fixed at 1. The comparison yielded Δχ2 = 21.27 with Δdf = 1, supporting discriminant validity between safety coaching and safety knowledge.
Based on the correlation analysis, safety participation was positively correlated with safety compliance (r = 0.80), resilience (r = 0.74), psychological contract (r = 0.75), and safety knowledge (r = 0.80). These results indicate that higher levels of safety compliance, resilience, psychological contract, and safety knowledge among construction workers are associated with increased safety participation.
Furthermore, safety compliance was positively correlated with resilience (r = 0.68) and safety knowledge (r = 0.72), suggesting that workers with greater resilience and safety knowledge tend to have higher safety compliance. The correlation between safety formalism and psychological contract was negative (r = −0.28), implying that a formalistic approach to safety regulations among construction workers may reduce their psychological contract. Lastly, safety coaching was strongly and positively correlated with safety knowledge (r = 0.83), indicating that a high perceived availability of safety coaching is associated with a higher level of safety knowledge.
For the purpose of testing the configural and metric invariance across two broad occupational categories, the sample was divided into manual and technical roles. The fit of the unconstrained model (testing configural invariance) yielded a CFI of 0.937. Subsequently, the regression coefficients were constrained to be equal across the two groups (testing metric invariance). The fit for this constrained model was CFI = 0.937. The change in CFI (ΔCFI) was 0.000, falling below the conventional threshold of 0.01. These findings indicate that the measurement model operates equivalently across the manual and technical respondents, confirming the establishment of both configural invariance and metric invariance.

4. Results

This study used the lavaan package in R to perform path analysis and hypothesis testing, estimating models with Unweighted Least Squares with Mean and Variance adjustment (ULSMV) [73]. The regression models included seniority and job position as control variables. The R-squared values for safety knowledge, psychological contract, compliance, and participation were 0.77, 0.10, 0.67, and 0.92, respectively. As shown in Table 4, safety formalism has a significant negative effect on the psychological contract, with a path coefficient of −0.29, thereby supporting H1. When construction workers perceive that organizational safety policies and regulations are not pragmatically implemented, their sense of a safety-related psychological contract decreases. Governmental agencies’ reliance on policy formalism often induces a corresponding tendency toward safety formalism within engineering units. When construction organizations perceive that the government does not genuinely prioritize worksite safety, they are less likely to faithfully enforce safety rules and procedures. Consequently, when workers observe that safety measures are not rigorously carried out, they perceive that the organization fails to fulfill its obligations, thereby weakening their psychological contract regarding safety.
Safety coaching has a significant positive effect on safety knowledge, with a path coefficient of 0.88, thereby supporting Hypothesis 2. This result is consistent with prior findings by Geller et al. [36] and Geller [37]. The more effort organizations devote to safety coaching and training, the greater the enhancement of construction workers’ safety knowledge. Proper instruction is necessary for workers regarding the use of personal protective equipment, working at heights, electrical safety, and crane operation safety, all of which require systematic organizational guidance.
Safety knowledge was found to have a significant positive effect on safety compliance and safety participation, with path coefficients of 0.45 and 0.36, respectively, thereby supporting Hypotheses 3 and 4. This result is consistent with the findings of Christian et al. [10] and Griffin and Neal [23]. When construction workers are well-acquainted with safety training, procedures, and regulations, their familiarity increases their willingness to comply with safety requirements and to participate in safety-related activities.
Resilience was found to have a significant positive effect on safety compliance and safety participation, with path coefficients of 0.45 and 0.32, respectively, thereby supporting Hypotheses 5 and 6. This result is consistent with the conclusions of Hollnagel [50] and Costella et al. [51]. In the face of the complex, dynamic, and highly uncertain environment of construction sites, workers with higher resilience are more willing to engage in safety compliance and safety participation.
The safety psychological contract was found to have a significant positive effect on safety participation, with a path coefficient of 0.13, thus supporting Hypothesis 7. This finding is consistent with the conclusions of Newaz et al. [58]. According to social exchange theory, when construction workers perceive that the organization fulfills its obligations toward safety, they increase their own commitment to safety [54]. Consequently, workers with a stronger safety psychological contract are naturally more willing to participate in safety activities.
This study justifies that safety compliance has a significant positive effect on safety participation, with a path coefficient of 0.37, thus supporting Hypothesis 8. This result is consistent with the conclusions of several previous studies [62,63]. When construction workers are willing to wear personal protective equipment and comply with safety regulations, they are also more inclined to participate in safety training sessions and prepare safety reports. The causal coefficient diagram is shown in Figure 2. The R2 values for psychological contract, safety knowledge, safety compliance, and safety participation were 0.08, 0.77, 0.67, and 0.91, respectively.
The indirect effect of safety coaching on safety participation through safety knowledge was 0.32 (p < 0.001), while the total effect of safety coaching on safety participation was 0.46 (p < 0.001). The indirect effect of safety coaching on safety compliance via safety knowledge was 0.39 (p < 0.001), and the total effect of safety coaching on safety compliance was also 0.39 (p < 0.001). The indirect effect of safety knowledge on safety participation through safety compliance was 0.16 (p = 0.004), whereas the total effect of resilience on safety participation was 0.49 (p < 0.001).

5. Discussion

This study used path analysis in structural equation modeling (SEM) to test the proposed hypotheses. This study reveals that safety formalism undermines construction workers’ psychological safety contract. When regulations are not rigorously implemented, workers perceive a breach of organizational obligations, weakening their psychological contract and reducing their perceived duty to act safely. These informal contracts still substantially influence safety participation. The findings of this study converge with previous research, affirming that construction workers’ perception of formalism is detrimental to actual site operations [77].
This study demonstrates that safety coaching and training contribute to the growth of workers’ safety knowledge. This finding is consistent with the conclusions drawn by Geller et al. [36] and Geller [37]. The systematic delivery of safety training in construction enhances workers’ safety knowledge (e.g., PPE, height, and crane protocols). This essential knowledge base is directly linked to, and thus expected to improve, workers’ safety compliance and participation on the construction site.
The findings show that safety knowledge is a significant antecedent to both safety compliance and safety participation. This finding is consistent with the previous research conducted by Christian et al. [10] and Griffin and Neal [23]. Workers must first understand safety regulations and knowledge before they can progress to the stages of safety compliance and participation. From another perspective, when workers understand the importance of safety knowledge in reducing workplace accidents, they are more willing to perform safety-related behaviors.
This study demonstrates that worker resilience can enhance both safety compliance and safety participation. The findings of this study are consistent with those reported by Hollnagel [50] and Costella et al. [51]. Facing complex and uncertain site conditions, highly resilient construction workers show superior adaptability and a greater propensity for safety compliance and participation. Resilience is essential for enabling correct judgments and effective responses in hazardous environments.
When construction workers perceive that their organizations are fulfilling their commitments to safety, such as by providing personal protective equipment and implementing safety training and coaching, a relationship of trust and reciprocity is established between the workers and the organization. This reciprocal relationship, in turn, makes workers more willing to engage in safety compliance and participation behaviors.
It is empirically supported that safety compliance is an antecedent to safety participation. The results of this study are comparable to those reported by Barbaranelli et al. [62] and Guo et al. [63]. When workers develop the habit of wearing personal protective equipment and adhering to safety protocols, they transition from merely following regulations to actively participating in safety-related activities. Encouraging workers to be compliant with safety is therefore a necessary step to promote their involvement in safety activities.

6. Conclusions

6.1. Theoretical Findings

One of the primary contributions of this study is the inclusion of safety formalism in the research on construction worker safety. Previously, little attention has been paid to how workers perceive the degree to which safety policies and regulations are implemented. It was often assumed that the mere existence of safety regulations and procedures could reduce accidents. However, these regulations and procedures are often only verified through government-mandated safety inspections, and whether these inspections are carried out in a timely manner is also subject to formalism. This study found that when workers’ perception of safety formalism is high, their psychological safety contract is diminished.
Another key contribution of this study is the establishment of a causal model that links safety formalism, safety coaching, worker resilience, psychological safety contract, and safety knowledge to both safety compliance and safety participation. Research that incorporates safety formalism into models of construction safety behavior is particularly rare. Research has shown that variations in safety formalism, safety coaching, worker resilience, psychological safety contracts, and safety knowledge all influence the degree of safety compliance and participation. Finally, this study validates the causal relationships between safety knowledge, cognition, personal attributes, and behavior.

6.2. Practical Implications

Organizations must ensure that on-site safety regulations are implemented in a practical manner. Construction site safety performance should be based on an outcome-oriented evaluation, with clear and measurable safety performance targets established. It is crucial to regularly review safety procedures for their appropriateness and feasibility. Regular safety implementation meetings should be held to allow workers to participate and offer safety suggestions. Simultaneously, a transparent safety reporting mechanism should be established to strengthen the psychological safety contract between workers and the organization.
Safety knowledge is the foundation for both safety compliance and safety participation. Targeted training on the use of personal protective equipment and regulations for working at heights should be provided for different types of hazards. The distribution of safety manuals alone is not sufficient to ensure that workers acquire safety knowledge. Only through hands-on training, drills, and examinations can workers truly grasp and apply the knowledge. This study justifies the importance of worker resilience for safety behavior on construction sites. Organizations should offer psychological support programs to assist workers after high-stress or traumatic events. Finally, by fostering a culture of teamwork and mutual support in safety reporting, organizations can effectively maintain workplace safety.
In terms of enhancing worker resilience, practical strategies can be implemented, such as conducting cross-trade briefings weekly. These sessions require foremen from various trades (e.g., electrical, masonry, rebar) to mutually communicate and outline the high-risk interfaces anticipated for the coming week. Furthermore, the establishment of a non-punitive culture is crucial, encouraging workers to proactively report near misses or conditions of personal fatigue without fear of retribution. To strengthen safety knowledge, construction sites should implement the Hazard Hunt initiative: Foremen should lead workers through different work zones weekly in a proactive “troubleshooting” activity. Workers are required to identify at least three potential hazards and propose corresponding mitigation strategies.

7. Study Limitations and Recommendations for Future Research

Due to time and funding constraints, this study only collected 99 samples of construction workers in Taiwan. The extrapolation of these findings to other regions and countries is somewhat limited. It is recommended that future researchers collect larger samples for verification if research time and funding permit. While standardized scales and statistical methods can identify certain causal relationships, it is recommended that future researchers supplement their work with qualitative analysis to explore the topic in greater depth. This study employed a self-report, mono-method design, and respondents may have been subject to social desirability bias; both constitute potential limitations. The study also faces endogeneity risks: safety compliance and safety participation may exhibit reverse causality, and some relevant variables may be omitted from the causal model.
Due to funding limitations, this study adopted a cross-sectional design. Future researchers could employ a longitudinal design to analyze the dynamic changes in construction safety over time. Furthermore, since this study’s sample was drawn exclusively from Taiwan, the generalizability of the findings may be limited. It is suggested that future research collect samples from more countries for comparative analysis to enhance the external validity of the results.

Author Contributions

Conceptualization, C.-J.H.; methodology, C.-J.H. and H.-T.L.; software, C.-J.H.; validation, C.-J.H., T.-L.W. and H.-T.L.; formal analysis, C.-J.H. and H.-T.L.; investigation, C.-J.H., T.-L.W. and H.-T.L.; resources, C.-J.H.; data curation, C.-J.H. and T.-L.W.; writing—original draft preparation, C.-J.H. and T.-L.W.; writing—review and editing, C.-J.H., T.-L.W. and H.-T.L.; visualization, C.-J.H.; supervision, C.-J.H.; project administration, C.-J.H. and T.-L.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The data presented in this study is available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
SCSafety coaching
SCOSafety compliance
PCPsychological contract
SFSafety formalism
SKSafety knowledge
REResilience
SPSafety participation
CMVCommon method variance
CFAConfirmatory factor analysis
PNFIParsimony normed fit index
CFIComparative fit index
NNFINon-normed fit index
GFIGoodness-of-fit index
SRMRStandardized root mean square residual
CRComposite reliability
AVEAverage variance extracted
HTMTHeterotrait–monotrait
MSVMaximum shared variance
ASVAverage shared variance

Appendix A

Table A1. Measurement items.
Table A1. Measurement items.
VariablesItems
Safety coaching“The organization educates workers about the importance of site safety,”
“The organization clearly explains site safety concepts,”
“The organization does not describe a vision for what safety looks like (R).”
“My supervisor systematically instructs subordinates on how to handle safety issues.”
“My supervisor sets an example by strictly complying with construction site safety regulations.”
Safety compliance“I perform my work tasks in a safe manner,”
“I use safety protective equipment while working,”
“I follow safety procedures while working.”
“Due to time constraints, I sometimes do not fully comply with the correct and safe work procedures (R).”
“Due to over-familiarity with the task, I sometimes do not fully comply with the correct and safe work procedures (R).”
Psychological contract“The organization invites workers to participate in safety decision-making meetings,”
“The organization listens to workers’ safety-related opinions,”
“The organization conducts an investigation when a site safety accident occurs,”
“The organization arranges regular safety training.”
Safety formalism“I feel that safety regulations and their actual implementation are not exactly the same,”
“I feel that safety laws on the construction site are sometimes difficult to fully enforce,”
“I feel that many safety systems are not easy to implement,”
“I feel there are discrepancies between construction site safety regulations and their current implementation status.”
Safety knowledge“I know how to carry out work on the jobsite safely,”
“I know how to use safety equipment and standard work procedures,”
“I know how to maintain or improve workplace health and safety,”
“I have the knowledge to reduce the risk of accidents in the workplace.”
Safety participation“I proactively carry out tasks to improve site safety,”
“I am willing to assist my colleagues when they are in danger on-site,”
“I am willing to put in extra effort to maintain site safety.”
“I am unwilling to execute tasks or engage in activities that would facilitate improvements in workplace safety (R).”
Resilience“I am able to cope with changes in the environment,”
“No matter what happens, I am able to handle it,”
“I can still achieve my goals even when facing difficulties.”

Appendix B

Table A2. Heterotrait–monotrait (HTMT) ratio of correlations.
Table A2. Heterotrait–monotrait (HTMT) ratio of correlations.
1234567
1. Safety compliance1
2. Safety participation0.891
3. Safety knowledge0.750.871
4. Safety coaching0.690.850.881
5. Psychological contract0.830.860.820.731
6. Resilience0.750.830.690.780.671
7. Safety formalism0.380.410.390.370.310.461

References

  1. Leung, M.Y.; Liang, Q.; Olomolaiye, P. Impact of job stressors and stress on the safety behavior and accidents of construction workers. J. Manag. Eng. 2016, 32, 04015019. [Google Scholar] [CrossRef]
  2. Aryee, S.; Hsiung, H.H. Regulatory focus and safety outcomes: An examination of the mediating influence of safety behavior. Saf. Sci. 2016, 86, 27–35. [Google Scholar] [CrossRef]
  3. Fang, D.; Wu, C.; Wu, H. Impact of the supervisor on worker safety behavior in construction projects. J. Manag. Eng. 2015, 31, 04015001. [Google Scholar] [CrossRef]
  4. He, C.; McCabe, B.; Jia, G. Effect of leader-member exchange on construction worker safety behavior: Safety climate and psychological capital as the mediators. Saf. Sci. 2021, 142, 105401. [Google Scholar] [CrossRef]
  5. Neal, A.; Griffin, M.A.; Hart, P.M. The impact of organizational climate on safety climate and individual behavior. Saf. Sci. 2000, 34, 99–109. [Google Scholar] [CrossRef]
  6. Mullen, J. Testing a model of employee willingness to raise safety issues. Can. J. Behav. Sci./Rev. Can. Des Sci. Du Comport. 2005, 37, 273–282. [Google Scholar] [CrossRef]
  7. Liu, Y.; Ye, L.; Guo, M. The influence of occupational calling on safety performance among train drivers: The role of work engagement and perceived organizational support. Saf. Sci. 2019, 120, 374–382. [Google Scholar] [CrossRef]
  8. Fugas, C.S.; Silva, S.A.; Meliá, J.L. Another look at safety climate and safety behavior: Deepening the cognitive and social mediator mechanisms. Accid. Anal. Prev. 2012, 45, 468–477. [Google Scholar] [CrossRef]
  9. Jiang, L.; Li, F.; Li, Y.; Li, R. Leader-member exchange and safety citizenship behavior: The mediating role of coworker trust. Work 2017, 56, 387–395. [Google Scholar] [CrossRef]
  10. Christian, M.S.; Bradley, J.C.; Wallace, J.C.; Burke, M.J. Workplace safety: A meta-analysis of the roles of person and situation factors. J. Appl. Psychol. 2009, 94, 1103–1127. [Google Scholar] [CrossRef]
  11. Goldenhar, L.M.; Williams, L.J.; Swanson, N.G. Modelling relationships between job stressors and injury and near-miss outcomes for construction labourers. Work Stress 2003, 17, 218–240. [Google Scholar] [CrossRef]
  12. Curcuruto, M.; Conchie, S.M.; Mariani, M.G.; Violante, F.S. The role of prosocial and proactive safety behaviors in predicting safety performance. Saf. Sci. 2015, 80, 317–323. [Google Scholar] [CrossRef]
  13. Teske, B.E.; Adjekum, D.K. Understanding the relationship between high reliability theory (HRT) of mindful organizing and safety management systems (SMS) within the aerospace Industry: A Cross-Sectional Quantitative Assessment. J. Saf. Sci. Resil. 2022, 3, 105–114. [Google Scholar] [CrossRef]
  14. Payne, S.C.; Dumlao, S.V.; Zhang, B.; Kang, J.; Mehta, R.K.; Sasangohar, F. The day-to-day stability of safety climate in the offshore oil and gas industry. J. Occup. Organ. Psychol. 2025, 98, e70051. [Google Scholar] [CrossRef]
  15. Luthans, F.; Avolio, B.J.; Avey, J.B.; Norman, S.M. Positive psychological capital: Measurement and relationship with performance and satisfaction. Pers. Psychol. 2007, 60, 541–572. [Google Scholar] [CrossRef]
  16. Luthans, F.; Luthans, K.W.; Luthans, B.C. Positive psychological capital: Beyond human and social capital. Bus. Horiz. 2004, 47, 45–50. [Google Scholar] [CrossRef]
  17. Luthans, F.; Norman, S.M.; Avolio, B.J.; Avey, J.B. The mediating role of psychological capital in the supportive organizational climate—Employee performance relationship. J. Organ. Behav. Int. J. Ind. Occup. Organ. Psychol. Behav. 2008, 29, 219–238. [Google Scholar] [CrossRef]
  18. Avey, J.B.; Luthans, F.; Youssef, C.M. The additive value of positive psychological capital in predicting work attitudes and behaviors. J. Manag. 2010, 36, 430–452. [Google Scholar] [CrossRef]
  19. Hobfoll, S.E. Conservation of resources: A new attempt at conceptualizing stress. Am. Psychol. 1989, 44, 513–524. [Google Scholar] [CrossRef] [PubMed]
  20. Braun, T.J.; Hayes, B.C.; DeMuth, R.L.F.; Taran, O.A. The development, validation, and practical application of an employee agility and resilience measure to facilitate organizational change. Ind. Organ. Psychol. 2017, 10, 703–723. [Google Scholar] [CrossRef]
  21. Luthans, F.; Youssef, C.M.; Avolio, B.J. Psychological Capital: Developing the Human Competitive Edge; Oxford University Press: Oxford, UK, 2006. [Google Scholar]
  22. Rego, A.; Sousa, F.; Marques, C.; e Cunha, M.P. Authentic leadership promoting employees’ psychological capital and creativity. J. Bus. Res. 2012, 65, 429–437. [Google Scholar] [CrossRef]
  23. Griffin, M.A.; Neal, A. Perceptions of safety at work: A framework for linking safety climate to safety performance, knowledge, and motivation. J. Occup. Health Psychol. 2000, 5, 347–358. [Google Scholar] [CrossRef] [PubMed]
  24. Wang, D.; Wang, X.; Xia, N. How safety-related stress affects workers’ safety behavior: The moderating role of psychological capital. Saf. Sci. 2018, 103, 247–259. [Google Scholar] [CrossRef]
  25. He, C.; Jia, G.; McCabe, B.; Chen, Y.; Sun, J. Impact of psychological capital on construction worker safety behavior: Communication competence as a mediator. J. Saf. Res. 2019, 71, 231–241. [Google Scholar] [CrossRef]
  26. Casey, T.; Turner, N.; Hu, X.; Bancroft, K. Making safety training stickier: A richer model of safety training engagement and transfer. J. Saf. Res. 2021, 78, 303–313. [Google Scholar] [CrossRef]
  27. Vinodkumar, M.N.; Bhasi, M. Safety management practices and safety behaviour: Assessing the mediating role of safety knowledge and motivation. Accid. Anal. Prev. 2010, 42, 2082–2093. [Google Scholar] [CrossRef]
  28. Milne, R.S. Mechanistic and organic models of public administration in developing countries. Adm. Sci. Q. 1970, 15, 57–67. [Google Scholar] [CrossRef]
  29. Shamsul Haque, M. Rethinking development administration and remembering Fred W. Riggs. Int. Rev. Adm. Sci. 2010, 76, 767–773. [Google Scholar] [CrossRef]
  30. Riggs, F.W. An ecological approach: The ‘Sala’ model. In Papers in Comparative Administration; Heady, F., Stokes, S., Eds.; Institute of Public Administration, University of Michigan: Ann Arbor, MI, USA, 1962; pp. 19–36. [Google Scholar]
  31. Tseng, W.Y.; So, W.Y. Inevitability of errors, justifiability of hidden rules: Behavior of performance appraisal under institutional constraints. J. Civ. Serv. 2017, 9, 79–107. [Google Scholar]
  32. Wu, T.-L.; Liu, H.-T. Causal model analysis of the effect of policy formalism, COVID-19 fear, social support and work stress on construction workers’ anxiety during the Epidemic. Buildings 2023, 13, 10. [Google Scholar] [CrossRef]
  33. Rodman, P.W. Development Administration: Obstacles, Theories and Implications for Planning; IIEP: Paris, France, 1968; Occasional Papers No. 2. [Google Scholar]
  34. Wiegand, D.M. Exploring the role of emotional intelligence in behavior-based safety coaching. J. Saf. Res. 2007, 38, 391–398. [Google Scholar] [CrossRef]
  35. Burke, M.J.; Salvador, R.O.; Smith-Crowe, K.; Chan-Serafin, S.; Smith, A.; Sonesh, S. The dread factor: How hazards and safety training influence learning and performance. J. Appl. Psychol. 2011, 96, 46–70. [Google Scholar] [CrossRef] [PubMed]
  36. Geller, E.S.; Perdue, S.R.; French, A. Behavior-based safety coaching: 10 guidelines for successful application. Prof. Saf. 2004, 49, 42–49. [Google Scholar]
  37. Scott Geller, E. Psychological science and safety: Large-scale success at preventing occupational injuries and fatalities. Curr. Dir. Psychol. Sci. 2011, 20, 109–114. [Google Scholar] [CrossRef]
  38. Hofmann, D.A.; Jacobs, R.; Landy, F. High reliability process industries: Individual, micro, and macro organizational influences on safety performance. J. Saf. Res. 1995, 26, 131–149. [Google Scholar] [CrossRef]
  39. Pandithawatta, S.; Ahn, S.; Rameezdeen, R.; Chow, C.W.K.; Gorjian, N. Systematic literature review on knowledge-driven approaches for construction safety analysis and accident prevention. Buildings 2024, 14, 3403. [Google Scholar] [CrossRef]
  40. Masten, A.S. Ordinary magic: Resilience processes in development. Am. Psychol. 2001, 56, 227–238. [Google Scholar] [CrossRef]
  41. Lazarus, R.S. From psychological stress to the emotions: A history of changing outlooks. Annu. Rev. Psychol. 1993, 44, 1–22. [Google Scholar] [CrossRef]
  42. Werner, E.E. Resilience in Development. Curr. Dir. Psychol. Sci. 1995, 4, 81–84. [Google Scholar] [CrossRef]
  43. Luthar, S.S.; Cicchetti, D.; Becker, B. The construct of resilience: A critical evaluation and guidelines for future work. Child Dev. 2000, 71, 543–562. [Google Scholar] [CrossRef]
  44. Liu, X.; Tang, W.; Xiao, R.; Zhou, W.; Huang, H.; Lin, R.; Tan, C.; Teng, X. Reducing psychological stress of Chinese adolescents by mindfulness-based intervention: The mediating role of resilience. In Child & Youth Care Forum; Springer: New York, NY, USA, 2023; pp. 351–370. [Google Scholar]
  45. Voon, S.P.; Lau, P.L.; Leong, K.E.; Jaafar, J.L.S. Self-compassion and psychological well-being among Malaysian counselors: The mediating role of resilience. Asia-Pac. Educ. Res. 2022, 31, 475–488. [Google Scholar] [CrossRef]
  46. Kocjan, G.Z.; Kavčič, T.; Avsec, A. Resilience matters: Explaining the association between personality and psychological functioning during the COVID-19 pandemic. Int. J. Clin. Health Psychol. 2021, 21, 100198. [Google Scholar] [CrossRef] [PubMed]
  47. Millear, P.; Liossis, P.; Shochet, I.M.; Biggs, H.; Donald, M. Being on PAR: Outcomes of a pilot trial to improve mental health and wellbeing in the workplace with the Promoting Adult Resilience (PAR) program. Behav. Change 2008, 25, 215–228. [Google Scholar] [CrossRef]
  48. Kermott, C.A.; Johnson, R.E.; Sood, R.; Jenkins, S.M.; Sood, A. Is higher resilience predictive of lower stress and better mental health among corporate executives? PLoS ONE 2019, 14, e0218092. [Google Scholar] [CrossRef] [PubMed]
  49. Hafeez, H.; Abdullah, M.I.; Zaheer, M.A.; Ahsan, Q. Organizational resilience process: Integrated model of safety culture. Organ. Manag. J. 2022, 19, 2–21. [Google Scholar] [CrossRef]
  50. Vanderhaegen, F. Erik Hollnagel: Safety-I and Safety-II, the past and future of safety management. Cogn Tech. Work 2015, 17, 461–464. [Google Scholar] [CrossRef]
  51. Costella, M.F.; Saurin, T.A.; de Macedo Guimarães, L.B. A method for assessing health and safety management systems from the resilience engineering perspective. Saf. Sci. 2009, 47, 1056–1067. [Google Scholar] [CrossRef]
  52. Blau, P. Exchange and Power in Social Life; Routledge: New York, NY, USA, 2017. [Google Scholar]
  53. Hui, C.; Lee, C.; Rousseau, D.M. Psychological contract and organizational citizenship behavior in China: Investigating generalizability and instrumentality. J. Appl. Psychol. 2004, 89, 311. [Google Scholar] [CrossRef]
  54. Rousseau, D. Psychological Contracts in Organizations: Understanding Written and Unwritten Agreements; Sage: London, UK, 1995. [Google Scholar]
  55. Xavier, I.M.; Jepsen, D.M. The impact of specific job stressors on psychological contract breach and violation. Hum. Factors Ergon. Manuf. Serv. Ind. 2015, 25, 534–547. [Google Scholar] [CrossRef]
  56. Choi, B.; Ahn, S.; Lee, S. Construction workers’ group norms and personal standards regarding safety behavior: Social identity theory perspective. J. Manag. Eng. 2017, 33, 04017001. [Google Scholar] [CrossRef]
  57. Walker, A. The development and validation of a psychological contract of safety scale. J. Saf. Res. 2010, 41, 315–321. [Google Scholar] [CrossRef] [PubMed]
  58. Newaz, M.T.; Davis, P.; Jefferies, M.; Pillay, M. Examining the psychological contract as mediator between the safety behavior of supervisors and workers on construction sites. J. Constr. Eng. Manag. 2020, 146, 04019094. [Google Scholar] [CrossRef]
  59. Hofmann, D.A.; Morgeson, F.P. Safety-related behavior as a social exchange: The role of perceived organizational support and leader–member exchange. J. Appl. Psychol. 1999, 84, 286. [Google Scholar] [CrossRef]
  60. DeJoy, D.M.; Della, L.J.; Vandenberg, R.J.; Wilson, M.G. Making work safer: Testing a model of social exchange and safety management. J. Saf. Res. 2010, 41, 163–171. [Google Scholar] [CrossRef]
  61. Saleem, M.S.; Isha, A.S.N.; Yusop, Y.M.; Awan, M.I.; Naji, G.M.A. The role of psychological capital and work engagement in enhancing construction workers’ safety behavior. Front. Public Health 2022, 10, 810145. [Google Scholar] [CrossRef]
  62. Barbaranelli, C.; Petitta, L.; Probst, T.M. Does safety climate predict safety performance in Italy and the USA? Cross-cultural validation of a theoretical model of safety climate. Accid. Anal. Prev. 2015, 77, 35–44. [Google Scholar] [CrossRef]
  63. Guo, B.H.; Yiu, T.W.; González, V.A. Predicting safety behavior in the construction industry: Development and test of an integrative model. Saf. Sci. 2016, 84, 1–11. [Google Scholar] [CrossRef]
  64. Pereira, E.; Ahn, S.; Han, S.; Abourizk, S. Finding causal paths between safety management system factors and accident precursors. J. Manag. Eng. 2020, 36, 04019049. [Google Scholar] [CrossRef]
  65. Lyubykh, Z.; Gulseren, D.; Turner, N.; Barling, J.; Seifert, M. Shared transformational leadership and safety behaviours of employees, leaders, and teams: A multilevel investigation. J. Occup. Organ. Psychol. 2022, 95, 431–458. [Google Scholar] [CrossRef]
  66. Vatankhah, S. Dose safety motivation mediate the effect of psychological contract of safety on flight attendants’ safety performance outcomes?: A social exchange perspective. J. Air Transp. Manag. 2021, 90, 101945. [Google Scholar] [CrossRef]
  67. Wu, T.C.; Chen, C.H.; Li, C.C. A correlation among safety leadership, safety climate and safety performance. J. Loss Prev. Process Ind. 2008, 21, 307–318. [Google Scholar] [CrossRef]
  68. Newaz, M.T.; Davis, P.; Jefferies, M.; Pillay, M. Using a psychological contract of safety to predict safety climate on construction sites. J. Saf. Res. 2019, 68, 9–19. [Google Scholar] [CrossRef] [PubMed]
  69. Shin, G.S.; Choi, K.S.; Jeong, K.S.; Min, Y.S.; Ahn, Y.S.; Kim, M.G. Psychometric properties of the 10-item Conner-Davidson resilience scale on toxic chemical-exposed workers in South Korea. Ann. Occup. Environ. Med. 2018, 30, 52. [Google Scholar] [CrossRef] [PubMed]
  70. Nunnally, J.C. Psychometric Theory; MacGraw-Hill: New York, NY, USA, 1967. [Google Scholar]
  71. Podsakoff, P.M.; MacKenzie, S.B.; Lee, J.Y.; Podsakoff, N.P. Common method biases in behavioral research: A critical review of the literature and recommended remedies. J. Appl. Psychol. 2003, 88, 879–903. [Google Scholar] [CrossRef]
  72. Fuller, C.M.; Simmering, M.J.; Atinc, G.; Atinc, Y.; Babin, B.J. Common methods variance detection in business research. J. Bus. Res. 2016, 69, 3192–3198. [Google Scholar] [CrossRef]
  73. Jöreskog, K.G.; Sörbom, D. LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language; Scientific Software International: Chicago, IL, USA, 1993. [Google Scholar]
  74. Hair Jnr, J.F.; Black, W.C.; Babin, B.J.; Anderson, R.E. Multivariate Data Analysis; Prentice-Hall, Inc.: Upper Saddle River, NJ, USA, 2010. [Google Scholar]
  75. Hulland, J. Use of partial least squares (PLS) in strategic management research: A review of four recent studies. Strateg. Manag. J. 1999, 20, 195–204. [Google Scholar] [CrossRef]
  76. Henseler, J.; Ringle, C.M.; Sarstedt, M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. J. Acad. Mark. Sci. 2015, 43, 115–135. [Google Scholar] [CrossRef]
  77. Wu, T.-L.; Chu, T.-F.; Liu, H.-T. Causal model analysis of the effect of policy formalism, equipment insufficiency and COVID-19 fear on construction workers’ job burnout, and insomnia during the epidemic. Buildings 2024, 14, 265. [Google Scholar] [CrossRef]
Buildings 15 04055 g001
Figure 1. Research framework.
Figure 1. Research framework.
Buildings 15 04055 g001
Buildings 15 04055 g002
Figure 2. Causal coefficient diagram. Note: *, **, and *** represent statistical significance at p < 0.05, p < 0.01, and p < 0.001, respectively.
Figure 2. Causal coefficient diagram. Note: *, **, and *** represent statistical significance at p < 0.05, p < 0.01, and p < 0.001, respectively.
Buildings 15 04055 g002
Table 1. Sample basic information.
Table 1. Sample basic information.
GenderPercentage (%)SeniorityPercentage (%)
Male66.71–3 years39.4
Female33.34–7 years21.2
AgePercentage (%)8–11 years9.1
18–29 years old28.312–15 years12.1
30–39 years old22.216 years and above18.2
40–49 years old27.3MarriagePercentage (%)
50 years old or older22.2Unmarried50
Job PositionPercentage (%)Married50
Laborer25.3Education LevelsPercentage (%)
Technician15.2Middle school (and below)2
Supervisor11.1High school14.1
Architect2Junior College9.1
Engineer18.2University51.5
Manager6.1Graduate School23.2
Administrator4
Other18.2
Table 2. Item loading and reliability.
Table 2. Item loading and reliability.
VariablesItemsLambdaLambda Z
Values		Composite ReliabilityCronbach’s Alpha
Safety coachingSC10.7812.450.920.92
SC20.8920.05
SC30.8919.56
SC40.8418.29
SC50.8313.85
Safety complianceSCO10.748.440.930.93
SCO20.9428.22
SCO30.9027.76
SCO40.8923.53
SCO50.8212.46
Psychological contractPC10.8013.050.870.87
PC20.7812.08
PC30.8413.22
PC40.7711.77
Safety formalismSF10.8311.400.920.92
SF20.9216.95
SF30.8621.89
SF40.8417.46
Safety knowledgeSK10.9029.460.950.95
SK20.9219.65
SK30.9126.96
SK40.9426.86
Safety participationSP10.7814.050.880.88
SP20.7814.99
SP30.7812.92
SP40.8825.62
ResilienceRE10.9119.320.880.88
RE20.7913.98
RE30.8317.53
Note: SC = safety coaching; SCO = safety compliance; PC = psychological contract; SF = safety formalism; SK = safety knowledge; RE = resilience; SP = safety participation.
Table 3. Square root of AVE and inter-correlations.
Table 3. Square root of AVE and inter-correlations.
1234567ASVMSVAVE
1. Safety compliance(0.85) 0.460.640.72
2. Safety participation0.80(0.81) 0.520.640.65
3. Safety knowledge0.720.80(0.92) 0.490.680.84
4. Safety coaching0.650.770.83(0.84) 0.460.680.71
5. Psychological contract0.750.750.750.66(0.79) 0.420.570.63
6. Resilience0.680.740.630.700.59(0.85) 0.410.550.72
7. Safety formalism−0.38−0.39−0.37−0.35−0.28−0.43(0.86)0.140.190.74
Note: The figures in parentheses indicate the square root of AVE of the study constructs. The lower left table on the diagonal is the Pearson correlation coefficient. MSV = maximum share variance, ASV = average share variance.
Table 4. Path coefficients.
Table 4. Path coefficients.
Causal PathPath
Coefficient		Standard
Error		95% Confidence
Interval		Z Valuep Value
LowerUpper
H1Safety formalism→Psychological contract−0.29 **0.10−0.49−0.09−2.880.004
H2Safety coaching→Safety knowledge0.88 ***0.030.820.9330.45<0.001
H3Safety knowledge→Safety compliance0.45 ***0.090.270.634.84<0.001
H4Safety knowledge→Safety participation0.36 ***0.090.190.534.16<0.001
H5Resilience→Safety compliance0.45 ***0.090.270.644.84<0.001
H6Resilience→Safety participation0.32 ***0.090.130.503.38<0.001
H7Psychological contract→Safety participation0.13 *0.060.020.242.290.022
H8Safety compliance→Safety participation0.37 ***0.100.170.573.64<0.001
Note: *, **, and *** represent statistical significance at p < 0.05, p < 0.01, and p < 0.001, respectively.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Huang, C.-J.; Wu, T.-L.; Liu, H.-T. Causal Model Analysis of the Impact of Formalism, Psychological Contract and Safety Coaching on Safety Compliance and Participation in Taiwan. Buildings 2025, 15, 4055. https://doi.org/10.3390/buildings15224055

AMA Style

Huang C-J, Wu T-L, Liu H-T. Causal Model Analysis of the Impact of Formalism, Psychological Contract and Safety Coaching on Safety Compliance and Participation in Taiwan. Buildings. 2025; 15(22):4055. https://doi.org/10.3390/buildings15224055

Chicago/Turabian Style

Huang, Chi-Jan, Tsung-Lin Wu, and Hsiang-Te Liu. 2025. "Causal Model Analysis of the Impact of Formalism, Psychological Contract and Safety Coaching on Safety Compliance and Participation in Taiwan" Buildings 15, no. 22: 4055. https://doi.org/10.3390/buildings15224055

APA Style

Huang, C.-J., Wu, T.-L., & Liu, H.-T. (2025). Causal Model Analysis of the Impact of Formalism, Psychological Contract and Safety Coaching on Safety Compliance and Participation in Taiwan. Buildings, 15(22), 4055. https://doi.org/10.3390/buildings15224055

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop