Next Article in Journal
Exploring Architectural Units Through Robotic 3D Concrete Printing of Space-Filling Geometries
Previous Article in Journal
Corrosion Characteristics and Flexural Performance of Carbonated Recycled Aggregate Concrete Beams in Corrosive Environments
 
 
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 1
10.3390/buildings15010059
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

The Impact of Accident Experience on Unsafe Behaviors of Construction Workers Within Social Cognitive Theory

by
Su Yang
1,
Lingyu Liu
1,
Ting Wang
1,
Yongqi Guo
2,
Yingmiao Qian
3,4 and
Huihua Chen
1,2,*
1
Anhui Institute of Real Estate and Housing Provident, School of Economics and Management, Anhui Jianzhu University, Hefei 230022, China
2
School of Civil Engineering, Central South University, Changsha 410075, China
3
School of Management Science and Engineering, Anhui University of Finance and Economics, Bengbu 233030, China
4
BIM Engineering Center of Anhui Province, Anhui Jianzhu University, Hefei 230022, China
*
Author to whom correspondence should be addressed.
Buildings 2025, 15(1), 59; https://doi.org/10.3390/buildings15010059
Submission received: 19 November 2024 / Revised: 18 December 2024 / Accepted: 22 December 2024 / Published: 27 December 2024
(This article belongs to the Section Construction Management, and Computers & Digitization)
Browse Figures
Versions Notes

Abstract

:
The construction industry’s poor safety is a global issue, with construction workers’ unsafe behaviors (CWUBs) identified as a major cause of accidents. Based on social cognitive theory (SCT) and using multiple regression analysis, this study categorizes accident experience (AE) into direct and indirect types, examining how each affects CWUBs and the roles of risk perception (RP), safety attitude (SA), and safety competence (SC) in these relationships. Utilizing a structured questionnaire completed by 334 valid respondents and analyzed through structural equation modeling (SEM), the study found that indirect experience (IE) significantly reduced CWUBs, with a standardized path coefficient of −0.364, while direct experience (DE) has a smaller impact, with a standardized path coefficient of −0.154, but a significant p. Furthermore, IE positively influenced RP, SA, and SC, explaining 66.8% of its total effect. This study offers a new framework for understanding how AE influences CWUBs, providing actionable insights for managers to implement effective strategies that reduce CWUBs on construction sites.

1. Introduction

Safety is a critical concern in the construction industry. Due to the complexity of construction environments and the large workforce involved, the industry has one of the highest casualty rates globally, second only to manufacturing [1]. Statistics reveal that the death rate in construction is 4.24 times higher than the average across all industries, with over 60,000 construction workers dying on job sites worldwide each year [2]. In China, the Housing and Urban–Rural Development Bureau reported 689 construction-related accidents in 2020, resulting in 794 deaths [3]. Similar trends are observed internationally: the U.S. Bureau of Labor Statistics (BLS) recorded 1008 construction worker fatalities in 2021, making up around 21% of all workplace fatalities [4]. In the UK, construction accounted for more than a third of all worker deaths in 2020 [5].
The Domino Theory of Accidents, one of the earliest models explaining accident causation, suggests that unsafe behaviors are primary contributors to accidents [6]. Subsequent research reinforces this idea, indicating that nearly 90% of accidents are closely linked to human behavior [7,8,9]. Unsafe behaviors in the construction industry are thus a major focus of research, with the goal of identifying their causes and designing interventions to reduce their occurrence. These behaviors are influenced by various personal, organizational, and managerial factors [10]. Among these, AE, defined as the influence of past incidents on present or future behaviors, is a significant factor shaping workers’ safety behaviors [11].
AE can be divided into direct and indirect types [12]. Direct experience (DE) refers to firsthand experiences of accidents, while IE involves learning about accidents through observing others or from reports and training sessions. Studies in sectors such as transportation and healthcare have shown that AE generally correlates with reduced CWUBs, even when accounting for intermediary factors [13,14]. In addition to AE, related factors considered to promote safety are RP, SA, and SC, which are essential to support safe practices in high-risk environments such as construction sites [15,16].
Building on these insights, the influence of AE on CWUBs has become a critical area for further investigation. Construction sites present unique hazards, and while prior research in other sectors indicates that AE plays a role in shaping workers’ safety behaviors [14], a comprehensive understanding of this influence in construction is still lacking. Specifically, the pathways through which AE both directly and indirectly affects CWUBs in construction remain unclear. Furthermore, limited attention has been given to how DE and IE may differ in their impacts on CWUBs.
The novelty of this research lies in its systematic categorization of AE into ED and IE forms, which have traditionally been treated as a single construct. This differentiation allows for a deeper understanding of the unique emotional, cognitive, and behavioral effects of these two types of AE. This study also examines their distinct effects on CWUBs through mediating factors such as RP, SA, and SC. By integrating SCT with empirical data from construction workers, this study develops and validates a theoretical framework that not only explains these mechanisms but also provides actionable insights for safety management in high-risk environments. This approach has led to significant advances in understanding the behavioral dimensions of construction safety, improving safety on construction sites.

2. Theoretical Mechanisms and Hypotheses

2.1. Social Cognitive Model

SCT, developed by Bandura, conceptualizes behavior as the outcome of dynamic interactions among individual, behavioral, and environmental factors [17]. This study employs SCT to investigate how AE influences CWUBs, with a focus on its impact through key mediators: RP, SA, and SC.
According to SCT, environmental factors play a significant role in shaping behavior by fostering observational learning and reinforcement. In this model, AE, comprising both DE and IE, serves as an environmental factor that heightens workers’ awareness of potential hazards by drawing on past experiences. SCT suggests that exposure to safety accidents, whether firsthand or observed, enhances workers’ ability to identify and evaluate risks, thereby guiding safer behavioral decisions [18].
SCT also emphasizes self-efficacy and self-regulation as core individual factors that shape how environmental inputs are processed. In this model, SA and SC align with these SCT concepts. SA reflects workers’ commitment to safety practices, corresponding to the SCT notion of self-efficacy, which motivates workers to address potential risks effectively. SC, representing the necessary skills and knowledge for managing safety hazards, aligns with the self-regulation process as workers adjust their behavior in response to perceived risks. SA and SC work together as self-regulatory mechanisms.
This SCT-based model proposes that AE, as an environmental factor, strengthens RP, which in turn enhances SA and SC, ultimately reducing CWUBs. By structuring these relationships within SCT’s framework, this study provides a comprehensive view of how AE, RP, SA, and SC interact to shape safety behaviors in high-risk construction environments [19,20]. More detailed factor analyses and hypothesis developments are discussed in the following sections.

2.2. Environmental Factors

AE is defined as the effect of past experiences on present and future behavior [14]. Since the early 20th century, the theory of accident propensity has suggested that AEs influence worker behavior [21]. In this study, AE is introduced as an environmental factor within the model to explain CWUBs. In this paper, AE is divided into two categories: DE and IE. DE refers to the experience of an accident in person; DE, as a first-hand experience, has a strong physical and psychological impact on construction workers [12]. The DE of construction workers seems to lead to more cautious behavior, such as actively wearing safety equipment and attending safety training. IE includes observing the accidents of others or receiving information through the media and education. Construction workers with IE, who have a clearer understanding of the potential consequences of accidents, will encourage themselves to actively learn and imitate colleagues who exhibit safe behaviors, thus reducing their unsafe behaviors in the workplace [22]. Based on this, we propose the following hypotheses about the impact of AE on unsafe behavior:
Hypothesis 1. 
DE has a negative effect on CWUBs.
Hypothesis 2. 
IE has a negative effect on CWUBs.
Previous studies have pointed out that workers’ AE can affect their own RP because people are accustomed to dealing with future risks based on past risk experience [23]. Scholars Rebecca Ferrer and William M. Klein [24] found that an individual’s specific experience would affect the level of RP. Personal experience helps to increase people’s RP to a particular threat. Construction workers who experience accidents directly tend to develop a stronger awareness of potential risks, as they retain clear memories of the accidents and their consequences. This awareness increases their readiness and ability to respond to similar situations in the future [25]. IE, although less immediate, also raises workers’ awareness of risk by allowing them to observe the effects of accidents on others, which reinforces their understanding of vulnerability and enhances their RP [26,27]. Based on this reasoning, we propose the following hypotheses regarding the effects of AE on RP:
Hypothesis 3. 
DE has a positive effect on RP.
Hypothesis 4. 
IE has a positive effect on RP.
Beyond its impact on RP, AE also plays an important role in enhancing workers’ SA and SC [28]. Through direct exposure to accidents, DE has a strong psychological impact, reinforcing the importance of safety and encouraging workers to take safety practices more seriously. This increased awareness translates into a more committed SA, as workers with DE are more likely to internalize safety protocols and develop a proactive attitude toward maintaining safety [29]. Furthermore, the rich experience that comes with DE often motivates workers to build specific skills that help them deal with similar risks more effectively in the future, thereby improving their overall competence in safety practices [12,13]. IE, while lacking the emotional impact of DE, also influences safety behaviors by enhancing awareness through observation. When workers witness accidents involving others, they recognize the importance of following safety practices, which strengthens their SA. Additionally, IE motivates workers to improve their SC by learning from others’ experiences and applying these lessons to prevent similar risks even if they have not personally encountered the dangers. Therefore, AE is an effective way to enhance SA and SC, and the following hypotheses are proposed:
Hypothesis 5. 
DE has a positive effect on SA.
Hypothesis 6. 
DE has a positive effect on SC.
Hypothesis 7. 
IE has a positive effect on SA.
Hypothesis 8. 
IE has a positive effect on SC.
From the SCT perspective, AE functions as an environmental factor that influences workers’ safety behaviors by engaging key SCT mechanisms. When workers experience or observe accidents, they gain insights into potential hazards, which heightens their RP. This awareness aligns with SCT’s concept of observational learning, where individuals adapt behaviors based on environmental cues. Increased RP then encourages workers to strengthen their SA, reflecting SCT’s self-efficacy principle, as they become more committed to maintaining safety standards. Additionally, AE supports the development of SC, aligning with SCT’s self-regulation mechanism, as workers actively adjust their skills to manage risks. Together, these processes illustrate how AE, through SCT’s core principles, fosters a proactive safety mindset that reduces CWUBs.

2.3. Personal Factors

In this study, individual factors are included in the model to explain CWUBs, focusing on SA and SC.
SA refers to how individuals perceive the importance of safety at work and includes their emotional responses and behavioral tendencies related to safety rules [30]. SA generally consists of three dimensions: cognitive, emotional, and behavioral tendencies [31]. Positive SA directly influences decision-making [32], encouraging workers to recognize hazards and take preventive actions, such as reporting unsafe conditions. Conversely, weak SA increases the likelihood of unsafe practices [33]. Research shows that workers with a strong commitment to safety are more likely to engage in safe behaviors, thereby reducing CWUBs. Therefore, we propose the following hypothesis:
Hypothesis 9. 
SA has a negative effect on CWUBs.
SC refers to the ability of construction workers to identify and manage hazards effectively, combining knowledge, skills, and personal values [34]. It is a key factor influencing CWUBs [35]. A lack of SC can lead to an increased likelihood of unsafe behaviors and accidents. Workers’ behaviors are influenced by their specific abilities [36]. Inadequate SC limits risk identification and safety decisions, increasing the likelihood of CWUBs [37]. Studies have consistently found that many construction workers lack adequate knowledge and competence [38] and that those with better skills and knowledge are more likely to adopt effective safety measures and reduce unsafe behaviors [39]. Hence, we propose the following hypothesis:
Hypothesis 10. 
SC has a negative effect on CWUBs.
In conclusion, SA and SC are critical individual factors affecting CWUBs, reflecting key concepts of SCT. A positive SA enhances workers’ perceptions of safety and motivates safe behaviors, aligning with SCT’s principle of self-efficacy. Meanwhile, SC involves the knowledge and skills necessary for effective safety management, corresponding to self-regulation, as workers with higher SC can better adjust their behaviors in response to risks. Improving both SA and SC is vital for enhancing safety outcomes in the construction industry.

2.4. Mediating Factor

RP refers to individuals’ subjective evaluations and intuitive judgments about risks [40]. It plays a crucial role in determining how workers respond to potential hazards in the workplace. In this study, we argue that AE significantly enhances workers’ RP by providing insights into the serious consequences of CWUBs. When workers experience or observe accidents, they become more aware of the potential dangers associated with their work. This heightened RP leads to a stronger SA. Workers who recognize the importance of safety are more likely to adopt positive attitudes toward safety practices. This connection suggests that as workers’ RP increases, so does their commitment to safety measures. In addition, increased RP influences SC. When workers are more aware of risks, they are motivated to improve their knowledge and skills related to safety management. This motivation drives them to seek training and apply their learning effectively, which enhances their ability to manage safety risks. Finally, a higher level of RP is associated with a decrease in CWUBs. Workers who understand the risks are more likely to take preventive measures and adhere to safety protocols. This proactive approach reduces the likelihood of engaging in unsafe behaviors, making them more competent in managing potential hazards. Therefore, we propose the following hypotheses:
Hypothesis 11. 
RP has a positive effect on SA.
Hypothesis 12. 
RP has a positive effect on SC.
Hypothesis 13. 
RP has a negative effect on CWUBs.
Therefore, RP serves as a critical mediating factor within the SCT framework, connecting AE to SA and SC while influencing CWUBs. Heightened RP, resulting from AE, enhances workers’ awareness of risks, leading to more positive SA and improved SC. This aligns with SCT’s principles, as increased awareness fosters self-regulation, empowering workers to make informed safety decisions.
The hypothesized conceptual framework is shown in Figure 1.

3. Research Methodology

3.1. Measurement

The research targeted construction workers from Hefei, Zhengzhou, Changsha, and Wuhan, China, to ensure representation from regions with high construction activity and diverse project types. The sample size of 500 workers was determined based on the principle of achieving statistical power for SEM, as well as practical considerations of data collection feasibility. Previous studies in similar contexts suggested a minimum of 300 valid responses for reliable SEM analysis, so a sample size of 500 is sufficient to ensure that there is enough margin to account for potentially unresponsive or incomplete data.
The survey included both male and female workers, reflecting the gender composition of the construction industry, though males constituted a higher proportion, accounting for 83.8%, due to the male-dominated nature of this workforce. Age criteria were broad, ranging from 18 to over 56 years, with 55.1% of respondents aged 45 years or older, highlighting the presence of experienced workers. The diversity of workers’ education levels was also taken into account, with 86.5% of the respondents having an education degree below university level, consistent with the typical demographic profile of construction workers in China. Their work experience was diverse; workers with less than five years of service accounted for 13.2% of the sample and those with more than 20 years of service accounted for 21.9%, ensuring the representation of different career stages.
This heterogeneity enhances the generality of the findings in different contexts. Trained interviewers were deployed to explain survey items to ensure respondents fully understood the questions, further improving the validity and reliability of the collected data. Ultimately, 426 responses were received, achieving an 85.2% raw response rate, with 334 valid questionnaires retained after screening, resulting in a valid response rate of 66.8% (Table 1).

3.2. Participants

The research was conducted in Hefei, Zhengzhou, Changsha, and Wuhan, China, where 500 questionnaires were distributed in person to construction workers. To enhance the response rate and ensure clarity of the survey questions, trained interviewers were present to explain the survey items to respondents as needed. A total of 426 questionnaires were returned, resulting in a raw response rate of 85.2%. After excluding incomplete responses, 334 valid questionnaires were obtained, yielding a valid response rate of 66.8%. To improve the validity and reliability of the collected data, the following measures were emphasized: (1) participants were assured of the confidentiality of their responses, ensuring that their data would not be disclosed to superiors or employers; (2) it was affirmed that the data would be used solely for research analysis.

3.3. Statistical Analysis

Data analysis was conducted using SEM, a powerful statistical technique that allows for the evaluation of complex relationships between observed and latent variables within a theoretical model. SEM was chosen for this study because it effectively combines confirmatory factor analysis and path analysis within a single framework, enabling a more nuanced understanding of the relationships among AE, RP, SA, SC, and CWUBs. This capability is particularly important given the multi-dimensional nature of the theoretical model and the need to account for the mediating roles of RP, SA, and SC. Unlike traditional regression analysis, which is limited to testing direct effects, SEM allows for the inclusion of latent variables and the estimation of indirect effects, thereby providing a deeper insight into the pathways through which AE influences CWUBs. Moreover, SEM’s ability to incorporate measurement error into the analysis ensures greater precision and reliability of the results. Before applying SEM, confirmatory factor analysis (CFA) was performed to assess the validity and reliability of the measurement instruments used for the constructs [48]. Model fit was evaluated using indices such as the chi-square to degree of freedom ratio (χ2/df), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA), and comparative fit index (CFI). Acceptable fit criteria included an χ2/df ratio below 5, CFI and TLI values above 0.90, and an RMSEA below 0.08 [49]. To ensure construct validity, convergent validity was assessed by checking that factor loadings and composite reliability exceeded 0.7. Discriminant validity was confirmed when the square root of the average variance extracted (AVE) for each construct was greater than its correlations with other constructs [50,51]. Internal consistency reliability was measured using Cronbach’s alpha, with values above 0.7 indicating adequate reliability [52]. The AMOS 26 software was applied to develop the SEM model and test proposed hypotheses. Finally, the PROCESS v2.16.3 macro [53] was used to analyze mediating effects among variables in the SEM framework. This macro allows for the assessment of indirect effects, helping to determine how changes in one variable influence another through mediators. Using PROCESS provides reliable estimates of mediation effects and confidence intervals for the analysis.

4. Results

4.1. Socio-Demographic Characteristics

Table 2 summarizes the socio-demographic characteristics of the respondents. Males constituted 83.8% of the participants, highlighting the predominance of men in the construction sector. A notable 55.1% of the respondents were over 45 years old, indicating a significant presence of older workers in the industry. Furthermore, 64.4% of the participants reported having more than 10 years of work experience, suggesting that many construction workers possess extensive industry experience. Additionally, only 10.5% of respondents held a bachelor’s degree or higher, reflecting the generally low education levels among construction workers.

4.2. Scale Reliability and Validity Test

The evaluation of the measurement model utilized four indices to assess model fit: χ2/df CFI, TLI, and RMSEA [48]. Table 3 presents the results, indicating that each index met the acceptable thresholds. This suggests that the measurement model provided an adequate fit to the data.
Cronbach’s alpha [52] was calculated to assess the internal consistency of each construct, with all constructs exceeding the 0.7 threshold (Table 4), demonstrating strong reliability. Convergent validity was confirmed when factor loadings and composite reliability (CR) exceeded 0.7, and average variance extracted (AVE) was above 0.5 [51]. Table 4 indicates that these criteria were met, supporting the convergent validity of the measurement model.
Discriminant validity was assessed by comparing the square root of the AVE for each construct with the correlations between constructs. Table 5 shows that the square roots of the AVEs for all constructs exceed the inter-construct correlations, confirming satisfactory discriminant validity.
In summary, the measurement model demonstrated adequate fit, strong internal consistency reliability, and both convergent and discriminant validity, confirming its appropriateness for use in SEM.

4.3. Structural Equation Model Test

The research model and its associated hypotheses were evaluated using SEM in AMOS 26. The same four model fit indices (χ2/df, CFI, TLI, and RMSE) were utilized to assess the proposed research model, as previously applied in the measurement model evaluation [48]. As shown in Table 6, all indices met the established criteria, indicating that the proposed model effectively captured the hypothesized relationships.
Following the confirmation of model fit, the hypothesis test results were evaluated, as shown in Table 7. Standardized path coefficients illustrate the relationships between latent variables. Except the RP → SC path, the remaining 12 hypothesized paths are statistically significant. Specifically, paths IE → CWUBs, DE → RP, IE → RP, DE → SA, and SA → CWUBs were highly significant at the p < 0.001 level, providing robust support for these relationships. Paths DE → SC, IE → SA, IE → SC, SC → CWUBs, and RP → SA were significant at the p < 0.01 level, indicating strong associations within these pathways. Paths DE → CWUBs and RP → CWUBs reached significance at the p < 0.05 level, further supporting their hypothesized effects. The RP → SC path was not supported, likely due to construction workers’ generally lower educational levels and limited training opportunities. Although higher RP may increase awareness of potential hazards, it does not necessarily translate into improved SC without sufficient knowledge and skill development resources. These results collectively confirm the hypothesized model’s structure as illustrated in Figure 2.

4.4. Mediation Analysis

In the mediation analysis, PROCESS v2.16.3 [53] was employed to investigate the mediating roles of RP, SA, and SC in the relationship between AE and CWUBs. The analysis was conducted at a 95% confidence level using 5000 bootstrap samples, ensuring robust statistical inference. The results reveal significant direct and indirect effects for both DE and IE on CWUBs, highlighting multiple mediation pathways that explain how AE influences CWUBs (Table 8 and Table 9).
Table 8 shows that DE’s total effect on CWUBs is −0.462, with 64.5% of this effect being direct and 35.5% occurring through indirect paths involving RP, SA, and SC. This suggests that DE not only directly reduces CWUBs but also achieves substantial impact through mediated pathways. Among these, the DE → RP → SA → CWUBs and DE → SA → CWUBs pathways are particularly important, contributing 14.3% and 8.9% of the total effect, respectively. These findings suggest that DE significantly reduces CWUBs by enhancing RP and SA, leading to safer behaviors. In addition to these main paths, other indirect pathways contribute to the overall effect: DE → RP → CWUBs (8.4%) and DE → SC → CWUBs (3.9%). Although individually smaller, these paths collectively reinforce DE’s impact on CWUBs. Overall, these results highlight that DE most effectively reduces CWUBs through the combined influence of RP, SA, or SC, underscoring the critical roles of these mediators in fostering safer work behaviors.
The results in Table 9 indicate that IE relies more on direct effects (66.8%) than DE, which has a larger proportion of indirect effects. This difference likely stems from the nature of IE, which, lacking the intense emotional impact of DE, influences CWUBs more directly. Among IE’s indirect pathways, IE → RP → SA → CWUBs accounted for 12.4% of the total effect and had the largest influence, followed by IE → SA → CWUBs, accounting for 8.3% of the total effect. These pathways suggest that IE is as effective as DE in reducing CWUBs by enhancing RP together with SA or SC. Pathways IE → SC → CWUBs and IE → RP → CWUBs contributed less, but still enhanced the overall impact of IE on CWUBs, similar to DE.

5. Discussion

This paper has effectively validated a mechanistic model that explores the impact of AE on CWUBs, drawing from the SCT. Additionally, it examines the influence of personal and environmental factors on CWUBs, particularly focusing on the mediating role of construction workers’ personal factors and RP between AE and CWUBs. By revealing the underlying mechanism of the link between AE and CWUBs, this study helps to develop safety policies, interventions, and strategies aimed at mitigating such behaviors. Furthermore, it offers practical recommendations to enhance construction safety.

5.1. Theoretical Implications

Occupational health and safety is an important area of concern in the construction industry [56]. Unlike previous studies on unsafe behavior of construction workers, this study takes AE as the starting point to systematically explore the impact path of AE on CWUBs. Specifically, taking SCT as the theoretical basis, this study proposes that CWUBs are formed by the comprehensive action of environmental factors and personal factors [57] and discusses the influence mechanism in detail, including both direct and indirect influence of AE on CWUBs. The empirical results showed that AE has a negative influence on CWUBs. Compared with workers without AE, workers with AE significantly reduced CWUBs, which is also consistent with previous research results in other fields [58].
AE impacts CWUBs through direct and indirect pathways, influencing both immediate and sustained safety behaviors. The direct pathway involves immediate compliance with safety protocols, reducing unsafe behaviors on-site. In contrast, the indirect pathway operates through enhancing factors like RP, SA, and SC, fostering long-term safety awareness. In the case of DE, although it has a significant direct effect, the proportion of indirect effects in its total impact is higher than in IE. The larger proportion of indirect effects in DE can be attributed to the nature of the experiences. DE often involves personal AE, which leads to emotional and cognitive processing. This internal processing enhances RP, SA, and SC over time, resulting in a stronger indirect effect. Workers gradually internalize safety practices, shaping their behavior in a more sustained way through these indirect pathways. In contrast, IE usually refers to observing the accidents of others or participating in organized safety education, which provides direct, clear guidance for avoiding risks. This leads to a stronger direct impact, as workers are prompted to comply with safety protocols quickly and rely less on internal cognitive shifts.
The overall effect of DE on CWUBs is smaller than that of IE, which can be attributed to the nature of DE. In the process of the questionnaire survey, we found that 73.8% of workers with DE had multiple minor scratches, cuts, and falls, but no serious injuries were caused. Only 16.2% of workers reported a serious accident involving a fracture, head injury, or other serious injury that resulted in hospitalization. The construction workers with DE mostly encountered minor accidents, which occur more frequently and have no serious consequences. Such experiences can lead to a “normalization effect” in which workers perceive these risks as manageable and become less vigilant over time [22]. At the same time, some workers who have experienced serious accidents have serious psychological trauma, which leads them to show avoidance behavior in the face of potential risks at work, instead of taking active safety measures [59]. DE influences security behavior primarily through complex, emotionally driven indirect pathways that enhance RP, SA, and SC. While these indirect effects are valuable for long-term safety awareness, they require multiple mediating steps, which may weaken the direct impact of DE on unsafe behavior change. In contrast, IE, which is often obtained by observing other people’s accidents or participating in safety incident reflection, provides a clear, systematic view of the risks, makes clear the serious consequences of accidents, directly enhances compliance with safety rules, and thus has a more direct overall effect in reducing CWUBs.
However, although IE has a larger total effect coefficient than DE on CWUBs, this does not necessarily indicate it is more critical for long-term safety. The total effect coefficient primarily measures the immediate impact of an experience on behavior but does not capture the depth or lasting influence. The deeper emotional impact of DE, due to personal experience with accidents, can lead to lasting changes in RP, SA, and SC. These cognitive shifts may result in more durable safety behaviors over time, even if the immediate effect appears smaller. In other words, DE’s emotional and cognitive impacts are crucial for cultivating long-term safety awareness and a sustained safety culture.
Another interesting finding is that the positive effect of RP on SC was not significant in the empirical results. This is due to the relatively low educational level of construction workers as a whole [57]. Survey results provide support for this explanation. As highlighted in Table 2, 61.7% of the participants had an education level of junior middle school or below, while only 10.5% had attained college-level education or higher. Such low educational attainment limits workers’ ability to comprehend and internalize complex safety-related information, which is crucial for developing SC. Although their RP may be improved after AE, the RP does not translate into a specific SC due to a lack of sufficient knowledge and skill training. The improvement of RP mainly increases their awareness of risk, while the skills required to actually perform safe operations rely on more systematic learning and practice [60], and the improvement of RP alone does not automatically increase workers’ SC. At the same time, because construction projects often have strict schedule requirements, workers often face greater time and performance pressure [61]. In this case, workers may be more inclined to complete work tasks and neglect the development of safety skills. Even if workers have a high RP, they may prioritize completing work goals rather than spending time improving SC, resulting in a weak or insignificant effect of RP on SC.

5.2. Practical Implications

Based on the results of this study, practical suggestions can be made to decrease CWUBs, thereby reducing the incidence of accidents on construction sites.
(1)
Strengthen safety awareness through immersive experience.
The study highlights that personal AE plays a significant role in reducing CWUBs, but minor incidents without severe consequences may lead to a “normalization effect”, where workers perceive risks as manageable and become less vigilant over time. To overcome this, it is crucial to enhance the emotional and cognitive impact of safety experiences. Construction companies can achieve this by incorporating modern technologies such as virtual reality (VR) and augmented reality (AR) to simulate realistic accident scenarios [35,62]. These technologies allow workers to experience the serious consequences of accidents in a controlled environment, helping them internalize emotional identification. This approach not only strengthens immediate compliance with safety protocols but also promotes long-term safety awareness. Additionally, displaying accident videos on-site can provide continuous reminders of safety risks, reinforcing SA and reminding workers of the potential consequences of unsafe behaviors.
(2)
Develop an accident case database to strengthen warning education.
Drawing from the study’s findings, which show that workers deepen their understanding of safety after experiencing accidents, companies can further enhance safety awareness by facilitating regular accident case-sharing and reflection sessions [63]. These sessions allow workers to discuss personal experiences and learn from one another, strengthening the SA of the team [25]. In addition, construction companies should create an internal database of accident cases, documenting typical incidents and their outcomes [64]. This database can serve as a valuable resource for ongoing safety training and education. Furthermore, organizing review meetings after the completion of a project to analyze potential risks and learn lessons can further improve the level of worker safety. By reviewing real-world examples, workers can internalize safety practices and better apply them in future projects, promoting a culture of continuous learning and reinforcing a long-term commitment to safety.
(3)
Develop a comprehensive reward and punishment system to promote safety compliance.
A key finding from the study is the importance of both immediate safety compliance and long-term safety culture. To foster both, construction companies should implement a balanced reward and punishment system [65]. This can include incentives such as “Safety Champion” or “Best Safety Team” awards to recognize and motivate individuals and teams who demonstrate strong safety practices. At the same time, a clear and transparent punishment system should be established to punish workers who show unsafe behaviors in their operations [66]. This system of punishment should be hierarchical—starting with warnings, followed by fines and more severe penalties for repeated violations. Such a system ensures that safety behaviors are both consistently reinforced and corrected, encouraging immediate adherence to safety protocols while fostering a long-term commitment to safety.

5.3. Limitations and Future Research Opportunities

While the study presents important results, it is necessary to acknowledge several limitations. First, the data for the questionnaire survey were collected through cross-sectional surveys, which may not fully capture the dynamic process of CWUBs. Longitudinal studies would provide a more comprehensive understanding of how CWUBs evolve over time and allow for a deeper exploration of the interrelationships between influencing factors. Second, the study’s data were gathered from mainland cities in China such as Hefei and Wuhan, which may limit the generalizability of the findings to other regions. Future studies should replicate this research in different geographic areas to assess the broader applicability of the results. Finally, this study focused on individual and environmental factors such as AE, SA, SC, and RP, and did not consider organizational factors such as organizational climate, which may also influence CWUBs. Further research should explore these organizational influences to obtain a more complete picture of safety behavior in the construction industry.

6. Conclusions

This study developed and validated a model grounded in SCT to explore how AE impacts CWUBs. It also verifies the role of key factors such as RP, SA, and SC in shaping safety behaviors. The findings show that AE can both directly and indirectly influence CWUBs. Directly, AE affects behavior adjustment, while indirectly, AE influences CWUBs by shaping RP, which in turn affects SA and SC. Analyzing data from 334 participants using SEM, the results highlight that IE has a more significant total effect on reducing CWUBs, with a value of −0.903, compared to the total effect coefficient of DE, which was −0.462. The total effect of DE on CWUBs is lower than IE due to factors such as the normalization effects. DE has a higher proportion of indirect effects compared to IE due to long-term emotional and cognitive impacts. Based on these findings, several practical recommendations can be made to reduce CWUBs and enhance safety on construction sites. First, construction companies should use VR and AR to simulate realistic accident scenarios, improving immediate compliance and long-term safety awareness. Second, an internal accident case database and regular case-sharing sessions can strengthen SA and IE. Lastly, a reward and punishment system should be implemented to reinforce safety behaviors and promote a lasting safety culture. Overall, the findings contribute to the body of knowledge on construction safety management and provide actionable insights for safety managers, workers, and other stakeholders.

Author Contributions

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

Funding

This research was funded by Anhui Real Estate and Housing Provident Fund Research Institute Open Fund (Grant No. 2023FDC01 of Huihua Chen); Anhui University Philosophy and Social Science Research Fund (Grant No. 2023AH040033 of Su Yang); BIM Engineering Center of Anhui Province Open Fund (Grant No. AHBIM2022KF01 of Yingmiao Qian); Domestic and International Study Visit and Training Program for Outstanding Young Talents in Universities (Grant No. gxfx2017055 of Su Yang).

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by School of Economics and Management, Anhui Jianzhu University (approval No. AHJZ20230412).

Informed Consent Statement

Informed consent for participation was obtained from all subjects involved in the study.

Data Availability Statement

The data can be requested from the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

CWUBs: Construction workers’ unsafe behaviors; AE: accident experience; DE: direct experience; IE: indirect experience; RP: risk perception; SA: safety attitude; SC: safety competence; SCT: social cognitive theory.

References

  1. Comu, S.; Kazar, G.; Marwa, Z. Evaluating the attitudes of different trainee groups towards eye tracking enhanced safety training methods. Adv. Eng. Inform. 2021, 49, 101353. [Google Scholar] [CrossRef]
  2. Guo, X.; Liu, Y.; Tan, Y.; Xia, Z.; Fu, H. Hazard identification performance comparison between virtual reality and traditional construction safety training modes for different learning style individuals. Saf. Sci. 2024, 180, 106644. [Google Scholar] [CrossRef]
  3. Workplace Accident Notification of Building and Municipal Engineering, Ministry of Housing and Urban-Rural Development. Available online: https://www.mohurd.gov.cn/gongkai/zc/wjk/art/2022/art_17339_768565.html (accessed on 21 May 2021).
  4. Yang, S.; Wang, T.; Li, H.; Liu, L.; Yao, W.; Ren, G. The Cross-Cutting Effects of Age Expectation and Safety Value on Construction Worker Safety Behavior: A Multidimensional Analysis. Buildings 2024, 14, 2290. [Google Scholar] [CrossRef]
  5. Health and Safety Executive. Workplace Fatal Injuries in Great Britain. Available online: https://www.ukata.org.uk/documents/294/Workplace_fatal_injuries_in_Great_Britain_2020.pdf (accessed on 1 July 2020).
  6. Mei, Y.; Huang, J.; Liu, J.; Jia, L. A Study of Factors Influencing Construction Workers’ Intention to Share Safety Knowledge. Buildings 2024, 14, 440. [Google Scholar] [CrossRef]
  7. Bao, Z.; Lu, W. Developing efficient circularity for construction and demolition waste management in fast emerging economies: Lessons learned from Shenzhen, China. Sci. Total Environ. 2020, 724, 138264. [Google Scholar] [CrossRef] [PubMed]
  8. Fu, H.; Tan, Y.; Xia, Z.; Feng, K.; Guo, X. Effects of construction workers’ safety knowledge on hazard-identification performance via eye-movement modeling examples training. Saf. Sci. 2024, 180, 106653. [Google Scholar] [CrossRef]
  9. Salminen, S.; Tallberg, T. Human errors in fatal and serious occupational accidents in Finland. Ergonomics 1996, 39, 980–988. [Google Scholar] [CrossRef] [PubMed]
  10. Zheng, X.; Wang, Y.; Chen, Y.; Zeng, Q.; Jin, L. Influence of safety experience and environmental conditions on site hazard identification performance. Buildings 2023, 13, 251. [Google Scholar] [CrossRef]
  11. Oah, S.; Na, R.; Moon, K. The Influence of Safety Climate, Safety Leadership, Workload, and Accident Experiences on Risk Perception: A Study of Korean Manufacturing Workers. Saf. Health Work 2018, 9, 427–433. [Google Scholar] [CrossRef]
  12. Terum, J.A.; Svartdal, F. Lessons learned from accident and near-accident experiences in traffic. Saf. Sci. 2019, 120, 672–678. [Google Scholar] [CrossRef]
  13. Lheureux, F.; Auzoult, L. When the social discourse on violation behaviours is challenged by the perception of everyday life experiences: Effects of non-accident experiences on offending attitudes and habits. Accid. Anal. Prev. 2016, 94, 89–96. [Google Scholar] [CrossRef]
  14. Gonçalves, S.M.P.; da Silva, S.A.; Lima, M.L.; Meliá, J.L. The impact of work accidents experience on causal attributions and worker behaviour. Saf. Sci. 2008, 46, 992–1001. [Google Scholar] [CrossRef]
  15. Gegenfurtner, A.; Lehtinen, E.; Säljö, R. Expertise differences in the comprehension of visualizations: A meta-analysis of eye-tracking research in professional domains. Educ. Psychol. Rev. 2011, 23, 523–552. [Google Scholar] [CrossRef]
  16. Zhang, L.; He, J. Critical factors affecting tacit-knowledge sharing within the integrated project team. J. Manag. Eng. 2016, 32, 04015045. [Google Scholar] [CrossRef]
  17. Bandura, A. Social cognitive theory: An Agentic Perspective. Annu. Rev. Psychol. 2001, 52, 1–26. [Google Scholar] [PubMed]
  18. Bandura, A. Human agency in social cognitive theory. Am. Psychol. 1989, 44, 1175. [Google Scholar] [CrossRef] [PubMed]
  19. Bandura, A. Self-Efficacy: The Exercise Of Control; W H Freeman/Times Books/Henry Holt & Co.: New York, NY, USA, 1997. [Google Scholar]
  20. Wang, Y.; Shen, C.; Zuo, J.; Rameezdeen, R. Same tune, different songs? Understanding public acceptance of mega construction projects: A comparative case study. Habitat Int. 2021, 118, 102461. [Google Scholar] [CrossRef]
  21. Cree, T.; Kelloway, E.K. Responses to occupational hazards: Exit and participation. J. Occup. Health Psychol. 1997, 2, 304. [Google Scholar] [CrossRef] [PubMed]
  22. Kouabenan, D.R. Occupation, driving experience, and risk and accident perception. J. Risk Res. 2002, 5, 49–68. [Google Scholar] [CrossRef]
  23. Fu, H.; Xia, Z.; Tan, Y.; Guo, X. Influence of cues on the safety hazard recognition of construction workers during safety training: Evidence from an eye-tracking experiment. J. Civ. Eng. Educ. 2024, 150, 04023009. [Google Scholar] [CrossRef]
  24. Xia, N.; Wang, X.; Griffin, M.A.; Wu, C.; Liu, B. Do we see how they perceive risk? An integrated analysis of risk perception and its effect on workplace safety behavior. Accid. Anal. Prev. 2017, 106, 234–242. [Google Scholar] [CrossRef] [PubMed]
  25. Becker, J.S.; Paton, D.; Johnston, D.M.; Ronan, K.R.; McClure, J. The role of prior experience in informing and motivating earthquake preparedness. Int. J. Disaster Risk Reduct. 2017, 22, 179–193. [Google Scholar] [CrossRef]
  26. Tong, R.P.; Zhao, H.; Zhang, N.; Li, H.W.; Wang, X.L.; Yang, H.Q. Modified accident causation model for highway construction accidents (ACM-HC). Eng. Constr. Archit. Manag. 2021, 28, 2592–2609. [Google Scholar] [CrossRef]
  27. Terpstra, T. Emotions, Trust, and Perceived Risk: Affective and Cognitive Routes to Flood Preparedness Behavior. Risk Anal. 2011, 31, 1658–1675. [Google Scholar] [CrossRef]
  28. Peng, Y.B.; Zhang, S.; Wu, P. Factors influencing workplace accident costs of building projects. Saf. Sci. 2015, 72, 97–104. [Google Scholar]
  29. Fang, D.; Zhao, C.; Zhang, M. A Cognitive Model of Construction Workers’ Unsafe Behaviors. J. Constr. Eng. Manag. 2016, 142, 04016039. [Google Scholar] [CrossRef]
  30. Loosemore, M.; Malouf, N. Safety training and positive safety attitude formation in the Australian construction industry. Saf. Sci. 2019, 113, 233–243. [Google Scholar] [CrossRef]
  31. Li, Y.L.; Wu, X.; Luo, X.W.; Gao, J.Q.; Yin, W.W. Impact of Safety Attitude on the Safety Behavior of Coal Miners in China. Sustainability 2019, 11, 21. [Google Scholar] [CrossRef]
  32. Basahel, A.M. Safety leadership, safety attitudes, safety knowledge and motivation toward safety-related behaviors in electrical substation construction projects. Int. J. Environ. Res. Public Health 2021, 18, 4196. [Google Scholar] [CrossRef]
  33. Deng, Y.L.; Guo, H.L.; Meng, M.M.; Zhang, Y.; Pei, S.S. Exploring the Effects of Safety Climate on Worker’s Safety Behavior in Subway Operation. Sustainability 2020, 12, 23. [Google Scholar] [CrossRef]
  34. Biggs, H.C.; Biggs, S.E. Interlocked projects in safety competency and safety effectiveness indicators in the construction sector. Saf. Sci. 2013, 52, 37–42. [Google Scholar] [CrossRef]
  35. Rahman, F.A.; Arifin, K.; Abas, A.; Mahfudz, M.; Cyio, M.B.; Khairil, M.; Ali, M.N.; Lampe, I.; Samad, M.A. Sustainable Safety Management: A Safety Competencies Systematic Literature Review. Sustainability 2022, 14, 17. [Google Scholar] [CrossRef]
  36. Li, S.Q.; Wu, X.Y.; Wang, X.Z.; Hu, S.H. Relationship between Social Capital, Safety Competency, and Safety Behaviors of Construction Workers. J. Constr. Eng. Manag. 2020, 146, 04020059. [Google Scholar] [CrossRef]
  37. Aggarwal, A.; Dhurkari, R.K. Association between stress and information security policy non-compliance behavior: A meta-analysis. Comput. Secur. 2023, 124, 11. [Google Scholar] [CrossRef]
  38. Lyu, S.; Hon, C.K.H.; Chan, A.P.C.; Wong, F.K.W.; Javed, A.A. Relationships among Safety Climate, Safety Behavior, and Safety Outcomes for Ethnic Minority Construction Workers. Int. J. Environ. Res. Public Health 2018, 15, 16. [Google Scholar] [CrossRef]
  39. Xiang, Q.T.; Ye, G.; Liu, Y.; Goh, Y.M.; Wang, D.; He, T.T. Cognitive mechanism of construction workers’ unsafe behavior: A systematic review. Saf. Sci. 2023, 159, 106037. [Google Scholar] [CrossRef]
  40. Slovic, P. Perception of risk. Science 1987, 236, 280. [Google Scholar] [CrossRef]
  41. Donald, I.; Canter, D. Psychological factors and the accident plateau. Health Saf. Inf. Bull. 1993, 215, 5–12. [Google Scholar]
  42. Siu, O.-L.; Phillips, D.R.; Leung, T.-W. Age differences in safety attitudes and safety performance in Hong Kong construction workers. J. Saf. Res. 2003, 34, 199–205. [Google Scholar] [CrossRef]
  43. 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]
  44. Gyekye, S.A.; Salminen, S. Are “good soldiers” safety conscious? an examination of the relationship between organizational citizenship behaviors and perception of workplace safety. Soc. Behav. Personal. Int. J. 2005, 33, 805–820. [Google Scholar] [CrossRef]
  45. Ali, F.; Rasoolimanesh, S.M.; Sarstedt, M.; Ringle, C.M.; Ryu, K. An assessment of the use of partial least squares structural equation modeling (PLS-SEM) in hospitality research. Int. J. Contemp. Hosp. Manag. 2018, 30, 514–538. [Google Scholar] [CrossRef]
  46. Fogarty, G.J.; Shaw, A. Safety climate and the theory of planned behavior: Towards the prediction of unsafe behavior. Accid. Anal. Prev. 2010, 42, 1455–1459. [Google Scholar] [CrossRef]
  47. Yule, S.; Flin, R.; Murdy, A. The role of management and safety climate in preventing risk-taking at work. Int. J. Risk Assess. Manag. 2007, 7, 137–151. [Google Scholar] [CrossRef]
  48. Kline, R. Principles and Practice of Structural Equation Modeling, 4th ed.; Guilford Press: New York, NY, USA, 2016. [Google Scholar]
  49. Hu, L.T.; Bentler, P.M. Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychol. Methods 1998, 3, 424–453. [Google Scholar] [CrossRef]
  50. Ab Hamid, M.R.; Sami, W.; Mohmad Sidek, M.H. Discriminant Validity Assessment: Use of Fornell & Larcker criterion versus HTMT Criterion. J. Phys. Conf. Ser. 2017, 890, 012163. [Google Scholar]
  51. Bagozzi, R.P. Evaluating Structural Equation Models with Unobservable Variables and Measurement Error: A Comment. J. Mark. Res. 1981, 18, 375–381. [Google Scholar] [CrossRef]
  52. Cronbach, L. Coefficient alpha and the internal structure of tests. Psychometrika 1951, 16, 297–334. [Google Scholar] [CrossRef]
  53. Hayes, A.F. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach.; Guilford Press: New York, NY, USA, 2013. [Google Scholar]
  54. Siegrist, M.; Arvai, J. Risk Perception: Reflections on 40 Years of Research. Risk Anal. 2020, 40, 2191–2206. [Google Scholar] [CrossRef] [PubMed]
  55. McDonald, R.P.; Ho, M.-H.R. Principles and practice in reporting structural equation analyses. Psychol. Methods 2002, 7, 64–82. [Google Scholar] [CrossRef]
  56. Fu, H.; Xia, Z.; Tan, Y.; Peng, Y.; Fan, C.; Guo, X. Fear Arousal Drives the Renewal of Active Avoidance of Hazards in Construction Sites: Evidence from an Animal Behavior Experiment in Mice. J. Constr. Eng. Manag. 2024, 150, 04024146. [Google Scholar] [CrossRef]
  57. He, C.Q.; Hu, Z.; Shen, Y.Z.; Wu, C.L. Effects of Demographic Characteristics on Safety Climate and Construction Worker Safety Behavior. Sustainability 2023, 15, 20. [Google Scholar] [CrossRef]
  58. Lindberg, A.K.; Hansson, S.O.; Rollenhagen, C. Learning from accidents—What more do we need to know? Saf. Sci. 2010, 48, 714–721. [Google Scholar] [CrossRef]
  59. Rispler, C.; Luria, G. Employee experience and perceptions of an organizational road-safety intervention—A mixed-methods study. Saf. Sci. 2021, 134, 105089. [Google Scholar] [CrossRef]
  60. Han, J.H.; Roh, Y.S. Teamwork, psychological safety, and patient safety competency among emergency nurses. Int. Emerg. Nurs. 2020, 51, 5. [Google Scholar] [CrossRef]
  61. Duan, P.S.; Zhou, J.L. Cascading vulnerability analysis of unsafe behaviors of construction workers from the perspective of network modeling. Eng. Constr. Archit. Manag. 2023, 30, 1037–1060. [Google Scholar] [CrossRef]
  62. Vignoli, M.; Nielsen, K.; Guglielmi, D.; Mariani, M.G.; Patras, L.; Peirò, J.M. Design of a safety training package for migrant workers in the construction industry. Saf. Sci. 2021, 136, 105124. [Google Scholar] [CrossRef]
  63. Xu, J.; Cheung, C.; Manu, P.; Ejohwomu, O. Safety leading indicators in construction: A systematic review. Saf. Sci. 2021, 139, 105250. [Google Scholar] [CrossRef]
  64. Yang, J.J.; Ye, G.; Xiang, Q.T.; Kim, M.; Liu, Q.J.; Yue, H.Z. Insights into the mechanism of construction workers’ unsafe behaviors from an individual perspective. Saf. Sci. 2021, 133, 105004. [Google Scholar] [CrossRef]
  65. Zhang, Z.T.; Guo, H.L.; Gao, P.Z.; Wang, Y.; Fang, Y.H. Impact of owners? safety management behavior on construction workers? unsafe behavior. Saf. Sci. 2023, 158, 105944. [Google Scholar] [CrossRef]
  66. Li, H.; Lu, M.J.; Hsu, S.C.; Gray, M.; Huang, T. Proactive behavior-based safety management for construction safety improvement. Saf. Sci. 2015, 75, 107–117. [Google Scholar] [CrossRef]
Buildings 15 00059 g001
Figure 1. Theoretical model.
Figure 1. Theoretical model.
Buildings 15 00059 g001
Buildings 15 00059 g002
Figure 2. The research model’s findings include standardized path coefficients, which are denoted by values adjacent to the arrows (A solid line signifies statistical significance, while a dotted line denotes a lack of significance).
Figure 2. The research model’s findings include standardized path coefficients, which are denoted by values adjacent to the arrows (A solid line signifies statistical significance, while a dotted line denotes a lack of significance).
Buildings 15 00059 g002
Table 1. Definitions and sorted items for constructs.
Table 1. Definitions and sorted items for constructs.
ConstructsItemsContentsReference
Accident experience
(AE)		AE1The number of minor bumps I’ve experienced in the last six months. [12,14]
AE2The number of times I have been hospitalized in my past careers.
AE3In the past six months, I have witnessed the number of safety accidents with people around me.
AE4I know how many times the news has reported security incidents in the past six months.
AE5After learning about all safety incidents, I reflect on the reasons why safety accidents occur.
Safety attitude
(SA)		SA1I know the importance of safety in the construction industry. [30,41,42]
SA2When I detect a safety hazard, I stop working first.
SA3Safety issues that I always pay attention to in my production operations.
SA4I work with special care when I’m not sure if it’s safe.
Safety competence
(SC)		SC1I am familiar with and understand how to use security equipment and standard construction practices. [36,43]
SC2I know how to get the job done in a safe way.
SC3I know what to do and to whom to report a potential hazard if I find a potential hazard in my workplace.
SC4I am aware of the risks inherent in my work and understand the necessary safety measures that should be taken.
SC5I am aware of my rights and responsibilities in occupational safety.
Risk perception
(RP)		RP1I often worry about cuts, stabbs, bumps, etc. while working. [44,45,46]
RP2I often worry about safety accidents such as electric shock, falls, object blows, mechanical injuries, etc.
RP3I often worry that heat, dust, or toxic gases at work will cause harm to my body.
RP4I think my job is a very dangerous job.
RP5I can identify the sources of hazards and hidden hazards in my workplace.
Construction workers’
unsafe behaviors
(CWUBs)		CWUBs1Do not wear necessary safety equipment. [43,47]
CWUBs2Failure to strictly comply with safe operating procedures and regulations.
CWUBs3Safety practices are ignored due to rush work.
CWUBs4Will not take the initiative to report to the leadership the potential safety hazards on the construction site.
CWUBs5Do not want to participate in safety education training and safety work meetings.
Table 2. Socio-demographic details of questionnaire respondents.
Table 2. Socio-demographic details of questionnaire respondents.
ItemsDetailsFrequencyPercent
GenderMale28083.8%
Female5416.2%
Age18–25133.9%
26–335115.3%
34–448625.7%
45–5513741%
≥564714.1%
EducationPrimary school and below4513.5%
Junior middle school16148.2%
Senior middle school9327.8%
College, bachelor degree and above3510.5%
Number of households120.6%
2113.3%
36619.8%
411032.9%
57622.8%
≥66920.7%
Work experience≤5 years4413.2%
5–10 years7522.5%
11–15 years8826.3%
16–20 years5416.2%
≥20 years7321.9%
Accident experience
(AE)		0288.4%
1–3 times10531.4%
4–6 times9829.3%
7–12 times7321.9%
≥13 times309.0%
Table 3. Result of model fit indices measurement model.
Table 3. Result of model fit indices measurement model.
Model Fit IndicesRecommended ValuesValuesReferences
χ2/df<51.646 [54]
RMSEA<0.080.044
CFI>0.90.973
TLI>0.90.969
Table 4. Results of convergent validity and internal consistency assessment.
Table 4. Results of convergent validity and internal consistency assessment.
ConstructItemMeanSDFactor LoadingAVECRCronbach’s Alpha
SCSC13.401.3360.7970.66730.92300.897
SC23.381.2650.845
SC33.441.1850.836
SC43.471.170.733
SC53.521.2780.781
SASA13.311.0640.7490.63450.89590.872
SA23.161.1440.900
SA33.231.0940.840
SA43.240.9190.697
AEAE13.281.0680.9370.72650.91310.925
AE23.161.1760.895
AE53.251.2590.861
AE33.510.8840.8160.74710.85430.753
AE43.121.1620.785
RPRP13.121.1070.7610.69820.90170.898
RP23.161.1320.895
RP33.131.0750.922
RP43.361.0950.750
RP53.300.9730.673
CWUBsCWUBs12.991.140.8480.67460.91110.938
CWUBs22.961.1960.896
CWUBs33.011.2250.905
CWUBs42.861.160.761
CWUBs52.961.2190.923
Table 5. Results of discriminant validity assessment.
Table 5. Results of discriminant validity assessment.
DPIPSASCRPCWUBs
DP0.7265
IP0.270.7471
SA0.4980.3260.6345
SC0.2720.2840.3120.6673
RP0.5010.3920.4370.2860.6982
CWUBs−0.378−0.439−0.538−0.352−0.3550.6746
Table 6. Result of model fit indices structural model.
Table 6. Result of model fit indices structural model.
Model Fit IndicesRecommended ValuesValuesReferences
χ2/df<51.546 [48,55]
RMSEA<0.080.040
CFI>0.90.978
TLI>0.90.975
Table 7. Result of hypothesis testing.
Table 7. Result of hypothesis testing.
HypothesisStandardized Path
Coefficient		p-ValueResult
H1: DE has a negative effect on CWUBs.−0.154*Supported
H2: IE has a negative effect on CWUBs.−0.364***Supported
H3: DE has a positive effect on RP.0.431***Supported
H4: IE has a positive effect on RP.0.349***Supported
H5: DE has a positive effect on SA.0.283***Supported
H6: DE has a positive effect on SC.0.207**Supported
H7: IE has a positive effect on SA.0.239**Supported
H8: IE has a positive effect on SC.0.279**Supported
H9: SA has a negative effect on CWUBs.−0.335***Supported
H10: SC has a negative effect on CWUBs.−0.157**Supported
H11: RP has a positive effect on SA.0.162**Supported
H12: RP has a positive effect on SC0.0750.276Not supported
H13: RP has a negative effect on CWUBs.−0.159*Supported
Note: * Correlation is significant at 0.05 level; ** Correlation is significant at 0.01 level; *** Correlation is significant at 0.001 level.
Table 8. Chain mediation effect of RP, SA, and SC in DE and CWUBs.
Table 8. Chain mediation effect of RP, SA, and SC in DE and CWUBs.
CoeffEffectiveness Degree
Total effects−0.462100%
Direct effects−0.29864.5%
Total indirect effects−0.16435.5%
Path1. DE → SA → CWUBs−0.0418.9%
Path2. DE → SC → CWUBs−0.0183.9%
Path3. DE → RP → CWUBs−0.0398.4%
Path4. DE → RP → SA → CWUBs−0.06614.3%
Table 9. Chain mediation effect of RP, SA, and SC in IE and CWUBs.
Table 9. Chain mediation effect of RP, SA, and SC in IE and CWUBs.
CoeffEffectiveness Degree
Total effects−0.903100%
Direct effects−0.62066.7%
Total indirect effects−0.28333.3%
Path1. IE → SA → CWUBs−0.0708.3%
Path2. IE → SC → CWUBs−0.0374.4%
Path3. IE → RP → CWUBs−0.0708.2%
Path4. IE → RP → SA → CWUBs−0.10612.4%
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

Yang, S.; Liu, L.; Wang, T.; Guo, Y.; Qian, Y.; Chen, H. The Impact of Accident Experience on Unsafe Behaviors of Construction Workers Within Social Cognitive Theory. Buildings 2025, 15, 59. https://doi.org/10.3390/buildings15010059

AMA Style

Yang S, Liu L, Wang T, Guo Y, Qian Y, Chen H. The Impact of Accident Experience on Unsafe Behaviors of Construction Workers Within Social Cognitive Theory. Buildings. 2025; 15(1):59. https://doi.org/10.3390/buildings15010059

Chicago/Turabian Style

Yang, Su, Lingyu Liu, Ting Wang, Yongqi Guo, Yingmiao Qian, and Huihua Chen. 2025. "The Impact of Accident Experience on Unsafe Behaviors of Construction Workers Within Social Cognitive Theory" Buildings 15, no. 1: 59. https://doi.org/10.3390/buildings15010059

APA Style

Yang, S., Liu, L., Wang, T., Guo, Y., Qian, Y., & Chen, H. (2025). The Impact of Accident Experience on Unsafe Behaviors of Construction Workers Within Social Cognitive Theory. Buildings, 15(1), 59. https://doi.org/10.3390/buildings15010059

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