Assessing the Impact of Occupational Stress on Safety Practices in the Construction Industry: A Case Study of Saudi Arabia

Abstract

Workplace health and safety issues have long plagued the construction industry. While safety efforts have traditionally focused on physical risks, increasing attention is being paid to mental health and work-related stressors, which can negatively affect both productivity and safety. In Saudi Arabia, the construction sector presents a unique context because of its highly diverse, multinational workforce. Workers of different nationalities often operate on the same job site, leading to potential communication barriers, cultural misunderstandings, and inconsistent safety practices, all of which may amplify stress and safety risks. This research aims to investigate the influence of work-related stressors on construction workers’ safety in Saudi Arabia and identify which stressors most significantly contribute to the risk of injury. A structured questionnaire was distributed to 349 construction workers across 16 job sites in Saudi Arabia. The survey measures ten key stressors identified in the literature, including job site demand, job control, job certainty, skill demand, social support, harassment and discrimination, conflict with supervisors, interpersonal conflict, and job satisfaction. Data were analyzed using logistic regression and Pearson correlation to examine relationships between stressors and self-reported injuries. The findings indicated that work-related stressors significantly predict workplace injury. While the first regression model showed a modest effect size, it was statistically significant. The second model identified job site demand and job satisfaction as the most influential predictors of injury risk. Work-related stressors, particularly high job demands and low job satisfaction, substantially increase the likelihood of injury among construction workers. These findings emphasize the importance of incorporating psychosocial risk management into construction safety practices in Saudi Arabia. Future studies should adopt longitudinal designs to explore causal relationships over time and include qualitative methods such as interviews to gain a deeper understanding. Additionally, factors such as nationality, organizational policies, and management style should be investigated to better understand their moderating effects on the stress–injury relationship.

1. Introduction

The construction industry has a long history of health and safety issues. Until recently, the primary focus of construction site safety experts and academics has been on the physical well-being of workers. However, recent research shows that mental health issues are increasingly leading to productivity loss and safety concerns [1]. More specifically, work-related stressors have been shown to increase the risk of injury in the construction industry [2]. Such factors as work overload [3,4], poor physical environments [3,5], interpersonal conflicts [2,6], organizational culture [7,8], and inadequate safety equipment [3,8] can cause acute psychological and physical stress. This stress, in turn, can lead to critical situations in the workplace, such as falls, being struck by an object, electrocutions, and being caught in/between objects [9,10,11]. This is compounded by the fact that construction work poses inherent risks, although workers are generally adept at recognizing hazards in their work environment [12,13]. While workplace injury and fatality rates have declined substantially in recent decades [14], they remain high compared with other economic sectors. For instance, the number of fatal injuries in the construction industry was 1061 in 2019 in the United States [15]. The fatal injuries in the construction industry are around 36% compared with the main industries in 2019/2020 in the United Kingdom [16], and it was the industrial sector with the highest rate of work-related injury or illness (59 per 1000 employed persons) in Australia [17].

Similar to the above countries, many construction accidents in Saudi Arabia result from insufficient safety measures on job sites. Statistics indicate that the construction sector accounts for approximately 48% of workplace injuries across all industries in Saudi Arabia [18,19,20,21,22,23,24,25,26,27].

Table 1. Construction injuries compared with total industrial injuries in Saudi Arabia.

Years	Total Injuries Among All Industries	Construction Industry Injuries	Construction (%)
2008	86,211	43,308	50.2
2009	75,487	37,527	49.7
2010	75,825	36,367	48.0
2011	65,656	31,048	47.3
2012	52,467	26,292	50.1
2014	67,087	35,587	53.0
2015	53,404	24,760	46.4
2017	42,361	19,362	45.7
2018	36,855	17,585	47.7
2019	30,547	11,931	39.1

shows the comparative analysis of construction and industrial injuries in Saudi Arabia from 2008 to 2019 based on open data published by the Saudi Social Insurance Agency [26,27].

In the current literature, the concept of “stress” remains ambiguously defined because of the absence of a universally accepted definition among researchers. However, certain academics represent stress as generalized physiological responses to external demands or stimuli [28]. Indeed, their discussion does not aim to provide direction to stress but to present a comprehensive overview of the stress phenomenon. In 1936, Hans Selye presented the term “stress” as it is known today, which was defined as “the nonspecific response of the body to any demand made upon it” [29]. Selye is widely recognized as a seminal figure in stress research, having provided one of the most comprehensive and enduring definitions of the concept, as outlined below:

“Associated with a great variety of essentially dissimilar problems, such as surgical trauma, burns, emotional arousal, mental or physical effort, fatigue, pain, fear, the need for concentration, the humiliation of frustration, the loss of blood, intoxication with drugs or environmental pollutants, or even with the kind of unexpected success that requires an individual to reformulate his lifestyle. Stress is present in the businessman under constant pressure; in the athlete straining to win a race; in the air traffic controller who bears continuous responsibility for hundreds of lives; in the husband helplessly watching his wife’s slow, painful death from cancer; in a racehorse, its jockey and the spectator who bets on them”

[30]

Stress can be a mental condition, work-related or personal, where mental stress represents some biochemical processes in the human body [31]. Work-related stress refers to a set of physiological, psychological, and behavioral reactions that arise in response to continuous negative influences experienced by individuals within an organization due to one or more stress-inducing factors. On the other hand, direct or indirect stressors, such as a family member with a serious illness or exposure to negative news, can also be considered personal stressors [32].

Indeed, the Saudi construction industry differs from other industries because of the significant presence of workers from diverse nationalities. While existing studies have explored work-related stress across various construction contexts, there is limited empirical evidence specifically examining how these stressors impact safety outcomes such as injury incidence, particularly in Saudi Arabia. Therefore, this study addresses this gap by identifying the most common stressors that influence construction workers’ safety on the job and by investigating the relationship between work-related stressors and self-reported injuries within the context of Saudi Arabia. Accordingly, the research questions addressed in this research are as follows:

(1)

To what extent do work-related stressors serve as reliable predictors of injury among construction workers in Saudi Arabia?

(2)

What are the key predictive factors of work-related stressors leading to workplace injuries?

(3)

Is there a statistically significant association between work-related stressors and self-reported injury incidence within the Saudi construction sector?

2. Literature Review

To address the safety and health-related concerns faced by construction workers, researchers have begun exploring the role of psychological antecedents influencing safety performance and mental well-being [33,34]. In this regard, stress has been extensively studied in other domains, and in recent years, it has begun to receive attention within the construction sector. According to researchers, the construction industry is widely recognized as a “high-risk setting” for poor mental health outcomes [35,36,37,38,39,40]. Construction workers frequently experience depression and chronic stress and often report reduced concentration and energy [39,41,42,43,44,45]. The nature of the job and the level of interaction with diverse parties can influence exposure to psychosocial stressors [43,44,46,47]. Based on a comprehensive literature review of 35 studies focused on construction stress, published between 1989 to 2025 within the construction and management domain, the authors have summarized and defined some of the most common sources of stress as follows: job site demand [2,3,6,9,48,49], job control [2,9,50], job certainty [2,48], interpersonal conflicts at work [34,49,51], role ambiguity [3], skill demand [2,49], social support [2,9,50], harassment and discrimination [2,48], supervisor conflicts at work [34], and job satisfaction [51]. Table 2 summarizes the most common sources of stress in the construction industry.

Job site demand refers to the amount of mental and physical effort people believe they will need to exert in their jobs [52]. According to the authors, the following are signs of a job demand stressor: work and role overload [49], emotional stress [53], physical fatigue [48], hours of exposure [2], a poor physical environment [48], a lack of goals [48], and task demands [3].

Regarding the relationship between job demand and workplace injury, Goldenhar et al. [2] found a low effect (r = 0.06) that was not statistically significant (p = 0.25). However, this contrasts with the findings of Leung et al. [48], who reported a significant relationship between job demand and worker injury (p < 0.05).

In the workplace, job control refers to the extent to which an employee can independently make decisions and manage their tasks [3,54]. This means that employees have more freedom to make decisions that could impact their tasks, work environment, motivation, and self-efficacy as a result of this freedom [2,9]. Both Goldenhar et al. [2] and Leung et al. [9] found that the relationship between the construction workers’ injuries and job control was significant.

Job certainty refers to how confident individuals feel about their employment status in their organization [2]. Stress at work is exacerbated by a lack of certainty, which affects the worker’s confidence and ability to perform their job effectively. In the absence of job certainty, a worker may be inclined to take unnecessary risks or make poor decisions that harm their safety and productivity [2,3]. The studies by Goldenhar et al. [2] and Leung et al. [9] both identified a significant relationship between job certainty and workplace injury.

The term “interpersonal conflicts at work” describes a wide range of negative interactions between coworkers, from insensitive behavior to heated disputes. Worker/individual resilience [34], role conflict [49], conflict with a coworker [34], and poor communication [51] are all stressors that the authors have assumed this stressor encompasses. As a subset of workplace conflict, this classification is entirely appropriate. Although this stressor was not found to be statistically significant in a study by Siu et al. [51], Chen et al. [49] found a significant relationship between interpersonal conflicts at work and workplace injury.

As another workplace stressor, the term “role ambiguity” refers to a situation where employees lack a clear understanding of their job responsibilities. Under intense work pressure, this uncertainty can lead individuals to make poor decisions because of confusion about their tasks and expectations [3]. In the existing literature, only a limited number of studies have explored the relationship between role ambiguity and safety performance.

Skill demand, meanwhile, is a stressor that arises from a worker’s inability to complete the work due to a lack of necessary skills and experience [49,55]. The authors considered the following factors as indicators of skill demand: over-compensation [2], underutilization [55], and a lack of safety knowledge [34]. In this regard, Chen et al. [49] identified a significant relationship between safety knowledge and workplace injury, while Goldenhar et al. [2] found a significant relationship between over-compensation and injury. However, they also found that the relationship between skill utilization and injury was not significant.

Social support refers to the relational support from coworkers, superiors, and the organization itself that can mitigate the psychological effects of a high-stress job [50]. The relationship between social support and workplace injury was found to be statistically insignificant in studies by Goldenhar et al. [2] and Leung et al. [9].

Discrimination and harassment are illegal acts perpetrated by individuals, organizations, communities, and governments targeting a specific person, gender, or group. Discrimination can be further defined as the systematic implementation of negative behaviors that target a specific individual or group of individuals, creating social, psychological, and physical barriers [56]. Any unfair treatment, improper conduct, exclusion, or retaliation in the workplace environment can be considered a form of discrimination [57]. The U.S. Equal Employment Opportunity Commission (EEOC) defines harassment, meanwhile, as “unwelcome conduct that is based on race, color, religion, sex (including sexual orientation, gender identity, or pregnancy), national origin, older age (beginning at age 40), disability, or genetic information” (2021). Leung et al. [3] note that inequitable reward systems or treatment of individuals are indicators of harassment and discrimination. The relationship between workplace harassment and discrimination, and injury, was found to be statistically insignificant in both the study by Goldenhar et al. [2] and that by Leung et al. [48].

Supervisor conflicts at work are defined as negative workplace interactions with one’s superiors, ranging from inconsistent treatment to blatantly disrespectful behavior [49]. On the other hand, a supervisor who refrains from such behaviors can play a crucial role in fostering a positive safety culture by ensuring that everyone is held accountable for their actions [58]. Chen et al. [49] found a statistically significant relationship between supervisor conflict and workplace injury.

Job satisfaction, finally, is a matter of striking a balance between the incentives offered by the workplace and the individual’s personal preferences for these incentives [59]. Individuals dissatisfied with their jobs may experience emotional exhaustion and distress. However, Siu et al. [51] claimed that the relationship between job satisfaction and workplace injury was not significant.

Table 2. Summary of common construction work-related stressors.

Construction Stressor	Definition	Authors
Job site demand	Refers to the amount of mental and physical effort people believe they will need to exert in their jobs.	[49,52,53]
job control	Refers to the extent to which an employee can independently make decisions and manage their tasks.	[3,54]
Job certainty	Refers to how confident individuals feel about their employment status in their organization.	[2]
Interpersonal conflicts at work	Describes a wide range of negative interactions between coworkers, from insensitive behavior to heated disputes. Worker/individual resilience, role conflict, conflict with a coworker, and poor communication are all stressors that the authors have assumed this stressor encompasses.	[34,49,51]
Role ambiguity	Refers to a situation where employees lack a clear understanding of their job responsibilities. Under intense work pressure, this uncertainty can lead individuals to make poor decisions due to confusion about their tasks and expectations.	[3]
Skill demand	Refers to a stressor that arises from a worker’s inability to complete the work due to a lack of necessary skills and experience.	[49,55]
Social support	Refers to the relational support from coworkers, superiors, and the organization itself that can mitigate the psychological effects of a high-stress job.	[50]
Discrimination and harassment	Refers to the illegal acts perpetrated by individuals, organizations, communities, and governments targeting a specific person, gender, or group, including psychological and physical barriers.	[56]
Supervisor conflicts at work.	Defined as negative workplace interactions with one’s superiors, ranging from inconsistent treatment to blatantly disrespectful behavior.	[49]
Job satisfaction	The matter of striking a balance between the incentives offered by the workplace and the individual’s personal preferences for these incentives.	[59]

Table 2. Summary of common construction work-related stressors.

Construction Stressor	Definition	Authors
Job site demand	Refers to the amount of mental and physical effort people believe they will need to exert in their jobs.	[49,52,53]
job control	Refers to the extent to which an employee can independently make decisions and manage their tasks.	[3,54]
Job certainty	Refers to how confident individuals feel about their employment status in their organization.	[2]
Interpersonal conflicts at work	Describes a wide range of negative interactions between coworkers, from insensitive behavior to heated disputes. Worker/individual resilience, role conflict, conflict with a coworker, and poor communication are all stressors that the authors have assumed this stressor encompasses.	[34,49,51]
Role ambiguity	Refers to a situation where employees lack a clear understanding of their job responsibilities. Under intense work pressure, this uncertainty can lead individuals to make poor decisions due to confusion about their tasks and expectations.	[3]
Skill demand	Refers to a stressor that arises from a worker’s inability to complete the work due to a lack of necessary skills and experience.	[49,55]
Social support	Refers to the relational support from coworkers, superiors, and the organization itself that can mitigate the psychological effects of a high-stress job.	[50]
Discrimination and harassment	Refers to the illegal acts perpetrated by individuals, organizations, communities, and governments targeting a specific person, gender, or group, including psychological and physical barriers.	[56]
Supervisor conflicts at work.	Defined as negative workplace interactions with one’s superiors, ranging from inconsistent treatment to blatantly disrespectful behavior.	[49]
Job satisfaction	The matter of striking a balance between the incentives offered by the workplace and the individual’s personal preferences for these incentives.	[59]

3. Methods

Figure 1 presents a flowchart outlining the research methodology. To address the research objectives, data were gathered through a systematically designed survey instrument distributed to construction workers across various sites in Saudi Arabia. The paper-based survey was hand-delivered and completed by the participants themselves. It included two sections: (1) questions about personal and job-related background, and (2) statements designed to explore different sources of stress that may affect how workers pay attention to safety on the job. The goal was to identify the types of stress, both personal and work-related, that could impact workers’ safety in the construction workplace. The stressor statements in part two were extracted from scales developed and adopted by Leung et al. [3], Leung et al. [9], Goldenhar et al. [2], and Siu et al. [60]. All responses were measured using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). To assess past injuries, a scale was used that categorized respondents based on whether they had ever been injured (0 = never injured, 1 = injured one or more times).

• The survey included several sets of statements intended to explore the impact of various work-related stressors on construction worker safety in Saudi Arabia. It also aimed to determine which stressors most frequently contribute to safety risks. These statements include:
• Three statements were presented to respondents to assess job site demand stressors: (1) “I am often asked to work in unsafe conditions during a typical workday. (2) I feel constant pressure at work [3]. (3) As a result of my job, I frequently feel angry or sad” [9]. Respondents were asked to indicate the extent to which they agreed with each of these statements.
• Two statements written to assess the role of ambiguity stressors; the statements are the following: (1) “My job duties are unclear, ambiguous, and ever-changing. (2) My supervisor communicates clearly what is expected from me on the job site” [3].
• Two statements are used to assess the stressors of job control: (1) “I have control over how to complete the tasks I am assigned” [9]. (2) “I have not been given the authority needed to carry out my tasks effectively” [3].
• One statement was used to assess the stressors of job certainty: (1) “I feel my position here is constantly under threat” [9].
• Two statements were used to assess the stressors of skill demands: (1) “I sometimes feel that I do have the skills to do the work I am asked to do.” (2) “The company provides me the opportunity to keep learning and improving at my job” [2].
• Two statements were used to assess the stressors of social support: (1) “My coworkers go out of their way to make my work life easier.” (2) “My coworkers are willing to make an extra effort to make my work life safer,” Goldenhar et al. [2].
• Two statements were used to assess the harassment and discrimination stressors: (1) “I have faced sexual harassment or discrimination at work” [2], (2) “I feel I am unfairly compensated for the work I do” [3].
• One statement was used to assess the supervisor’s conflicts at work stressors: (1) “I have a good relationship with my supervisor” [3].
• Two statements were used to assess the interpersonal conflicts at work: (1) “I have a good relationship with my coworkers” [3]. (2) “I have trouble communicating with other workers or supervisors on the job site”.
• Two statements were written to assess job satisfaction stressors: (1) “I like the work I do. (2) I take pride in the work I do” [60].

The questionnaire was initially developed in English and subsequently translated into six additional languages (i.e., Urdu, Arabic, Hindi, Nepali, Filipino, and Bengali) by certified translation agencies to accommodate the linguistic diversity of participants. All research materials, including the survey instrument, informed consent documents, and participant communication letters, received ethical approval from the Institutional Review Board (IRB) at Western Michigan University. Before the main data collection, a pilot survey was conducted with a panel of experts and a small group of construction workers to assess the clarity, relevance, and validity of the survey instrument. All questionnaire items were pre-tested during the pilot phase by administering the draft survey instrument to a small group of construction workers with backgrounds similar to those of the target sample. The responses were analyzed against the planned survey instrument analysis to ensure item clarity, logical flow, and measurement consistency. No wording adjustments were necessary, and no structural changes to the constructs or measurement scales were needed.

Over a two-month period, 349 male construction workers from 16 randomly selected construction sites from an official registry in the eastern region of Saudi Arabia participated in this study. The choice of workers within each randomly selected site was based on convenience sampling, which focused on their availability and willingness to take part during the survey period. This approach was chosen because of site access restrictions and the need to recruit participants who were available and willing to participate during the survey period. However, it is important to note that, similar to other self-reported survey studies, the responses in this research may be influenced by recall bias or the tendency to respond in a socially desirable manner. To mitigate these potential issues, the survey was conducted anonymously, and participants were informed that their input would remain confidential and solely used for academic purposes. Furthermore, all questionnaire items were pre-tested during the pilot phase and carefully formulated to enhance clarity and reduce misinterpretation. After collecting the completed surveys, the researchers used logistic regression models to explore which stress factors affected worker safety. They also used Pearson’s correlation analysis to study the relationship between work-related stress and self-reported injuries.

4. Results

Table 3. Participant’s profile.

Respondents Profile		Frequency	Percentage
Age	18–24	37	10.6
	25–34	116	33.2
	35–44	115	33
	45–54	70	20.1
	55–64	10	2.9
	65 or older	1	0.3
	Total	349	100
Years of experience in construction	0–2	35	10
	3–5	81	23.2
	6–8	90	25.8
	9–11	74	21.2
	12–14	35	10
	15 or more	34	9.7
	Total	349	100
Years of experience in Saudi Arabia	0–2	53	15.2
	3–5	95	27.2
	6–8	87	24.9
	9–11	59	16.9
	12–14	38	10.9
	15 or more	17	4.9
	Total	349	100

presents the demographic profile of the participants, including age, total years of construction experience, years of experience within the Saudi construction industry, and marital status. The largest age group was 25–34 years, comprising 33.2% of the sample, followed closely by those aged 35–44 (33%), 45–54 (20.1%), and 18–24 (10.6%). Regarding construction experience, 25.8% of participants reported having 6 to 8 years of experience, making it the most common range. Similarly, 27.2% of respondents indicated that they had lived in Saudi Arabia for 3 to 5 years. Of the total 349 participants, 260 (74.5%) were married.

As shown in

Table 4. Frequencies and percentages of respondents by nationality.

Nationality	Frequency	Percentage
Bangladesh	120	34.4
India	78	22.3
Pakistan	64	18.3
Nepal	35	10.0
Egypt	22	6.3
Philippines	20	5.7
Sudan	7	2.0
Syria	2	0.6
Saudi	1	0.3
Total	349	100.0

, the majority of participants were Bangladeshi nationals (34.4%, n = 120), followed by individuals from India (22.3%, n = 78) and Pakistan (18.3%, n = 64).

Table 5. Injury frequencies and percentages among participants.

How Many?	Frequency	Percentage
Never	159	45.6
1–2	141	40.4
3–4	33	9.5
5–6	4	1.1
7–8	5	1.4
9+	7	2.0
Total	349	100.0

displays self-reported injury data, with 159 participants (45.6%) stating they had never been injured, while 141 (40.4%) reported having experienced one or two injuries.

The second part of the survey comprised 19 items addressing various work-related stressors. The responses to these items were systematically analyzed to identify patterns and trends in the association between work-related stressors and the occurrence of workplace injuries. In modern research, regression techniques are frequently employed to describe associations between a response variable and one or more explanatory variables. Often, the outcome variable is categorical, with a limited set of possible values. For analyzing data of this nature, the logistic regression model is the most commonly used [61,62]. Given the binary nature of the outcome variable, this study utilized logistic regression analysis to examine the relationship between work-related stressors and workplace injury. Both the dependent and independent variables were categorical. To examine the research questions as shown below, statistical analyses were performed using IBM SPSS Statistics software: version 30.

(1)

To what extent do work-related stressors serve as reliable predictors of injury among construction workers in Saudi Arabia?

As an initial step, the mean score for each of the 19 survey items was calculated based on responses from 349 participants. The dependent variable in this analysis was self-reported injury, while the independent variable was the aggregated average score of work-related stressors. This analytical approach, referred to as ‘Model 1,’ is represented mathematically in Equation (1).

$$\begin{gathered} \mathrm{Model}1:\mathit{log}\left({\displaystyle \frac{injury}{\left(1-injury\right)}}\right)={\beta }_0+{\beta }_1\left(AverageofWork- \\ RelatedStressorsscore\right) \end{gathered}$$

(1)

As shown in

Table 6. Omnibus tests of model coefficients.

		Chi-Square	df	Sig.
Step 1	Step	7.699	1	0.006
	Block	7.699	1	0.006
	Model	7.699	1	0.006

, the overall logistic regression model was statistically significant, χ²(1) = 7.699, p < 0.05. Furthermore, based on the results in

Table 7. Model summary.

Step	−2 Log Likelihood	Cox and Snell R Square	Nagelkerke R Square
1	473.360 a	0.022	0.029

^a Estimation terminated at iteration number 3 because parameter estimates changed by less than 0.001.

, the model accounted for approximately 2.2% of the variance in injury outcomes, as indicated by Cox’s R². The Cox measure was reported instead of Nagelkerke’s R² because of the continuous nature of all independent variables. This method is suitable for models that do not include categorical predictors. However, the model’s Nagelkerke R² value was modest, implying that although the model achieved statistical significance, it accounts for only a limited proportion of the variation in self-reported injuries. This outcome points to the possible influence of additional factors not included in the model. Nonetheless, the Omnibus test of model coefficients yielded a statistically significant result, as shown in Table 6 (χ² = 7.699, df = 1, p = 0.006), confirming that the model offers a better fit than the null model. Variable inclusion was guided by their theoretical importance as supported by previous studies, and their relationships with the outcome were preliminarily evaluated using bivariate analysis. As presented in

Table 8. Classification table ^a.

	Observed		Predicted
			Injury		Percentage Correct
			0.00	1.00
Step 1	Injury	0.00	39	120	24.5
		1.00	37	153	80.5
	Overall Percentage				55.0

^a The cut value is 0.500.

, the logistic regression model demonstrated an overall classification accuracy of 55%, successfully identifying 80.5% of construction workers as being at risk of injury associated with work-related stressors. Based on the mean scores, all work-related stressors emerged as statistically significant predictors of workplace injury (Wald = 7.37, p < 0.05), as presented in

Table 9. Variables in the equation.

		B	S.E.	Wald	df	Sig.	Exp(B)	95% C.I. for EXP(B)
								Lower	Upper
Step 1 a	AVERAGE	0.660	0.243	7.377	1	0.007	1.934	1.202	3.113
	Constant	−1.883	0.767	6.033	1	0.014	0.152

^a Variable(s) entered in Step 1: AVERAGE.

. Additionally, a higher average score of work-related stressors was associated with a 1.934-fold increase in the likelihood of reporting an injury. However, as shown in

Table 10. Correlations.

		Injury	Average Score of Stressors
Injury	Pearson Correlation	1	0.148 **
	Sig. (2-tailed)		0.006

** Correlation is significant at the 0.01 level (2-tailed).

, the strength of the relationship between average stressor scores and injury occurrence was relatively weak.

(2)

What are the key predictive factors of work-related stressors leading to workplace injuries among construction workers in Saudi Arabia?

(3)

Is there a statistically significant relationship between work-related stressors and self-reported injury rates in Saudi Arabia’s construction industry?

In addressing the second research question, a key step in Model 2 involved calculating the average score for each of the ten identified stressors. For instance, the ‘job site demand’ stressor was represented by three survey items, each rated on a 5-point Likert scale. For each respondent, an average score across these three items was computed, followed by calculating the overall mean score for that specific stressor. This procedure was repeated for all ten stressors. In this model, the ten work-related stressors served as independent variables, while self-reported injury was the dependent variable. Model 2 is represented mathematically in Equation (2).

$$\begin{gathered} \mathrm{Model}2:\log \left({\displaystyle \frac{injury}{\left(1-injury\right)}}\right)={\beta }_0+{\beta }_1\left(jobsitedemand\right){+\beta }_2\left(jobAmbiguity\right)+ \\ {\beta }_3\left(jobcontrol\right){+\beta }_4\left(certainty\right)+{\beta }_5\left(skilldemand\right){+\beta }_6\left(socialsupport\right)+ \\ {\beta }_7\left(harassmentanddiscrimination\right){+\beta }_8\left(supervisorconflict\right)+ \\ {\beta }_9\left(interpersonalconflict\right)+{\beta }_{10}\left(Satisfaction\right) \end{gathered}$$

(2)

As presented in

Table 11. Tests of model significance.

	Chi-Square	df	Sig.
Model	28.227	10	0.002

, the logistic regression model was statistically significant, χ²(10) = 28.227, p < 0.05. Additionally,

Table 12. Model summary.

Step	−2 Log Likelihood	Cox and Snell R Square	Nagelkerke R Square
1	452.832 a	0.078	0.104

^a The cut value is 0.500.

shows that the model accounted for approximately 7% of the variance in injury outcomes, as indicated by Cox’s R². The Cox value was reported instead of Nagelkerke’s R², given that all independent variables were continuous. As shown in

Table 13. Classification table.

	Observed		Predicted
			Injury		Percentage Correct
			0.00	1.00
Step 1	Injury	0.00	92	67	57.9
		1.00	58	132	69.5
	Overall Percentage				64.2

, the model correctly classified 64.2% of all cases and identified 69.5% of construction workers who reported being at risk of injury because of work-related stress. Among the ten stressors, job site demand emerged as the most significant predictor of workplace injury (Wald = 17.40, p < 0.001), followed by job satisfaction (Wald = 5.24, p = 0.022). The remaining predictors did not show a statistically significant contribution to the model (see

Table 14. Variables in the equation.

		B	S.E.	Wald	df	Sig.	Exp(B)	95% C.I. for EXP(B)
								Lower	Upper
Step 1 a,b	Job site demand	0.675	0.162	17.393	1	<0.001	1.964	1.430	2.697
	Role ambiguity	−0.023	0.157	0.022	1	0.883	0.977	0.718	1.330
	Job control	0.018	0.156	0.013	1	0.909	1.018	0.750	1.382
	Job certainty	−0.116	0.115	1.018	1	0.313	0.890	0.711	1.116
	Skill demand	−0.121	0.156	0.608	1	0.436	0.886	0.653	1.201
	Social Support	0.114	0.153	0.556	1	0.456	1.121	0.830	1.514
	H&D	0.046	0.133	0.120	1	0.729	1.047	0.807	1.360
	SCW	−0.223	0.165	1.824	1	0.177	0.800	0.578	1.106
	ICW	−0.173	0.202	0.730	1	0.393	0.841	0.566	1.251
	Job satisfaction	0.396	0.173	5.232	1	0.022	1.485	1.058	2.085
	Constant	−1.434	0.861	2.770	1	0.096	0.238

^a The cut value is 0.500. ^b Note: H&D = Harassment and Discrimination, SCW = Supervisor Conflicts at Work, and ICW = Interpersonal Conflicts at Work.

). An increase in the job site demand score was associated with a 1.964-fold increase in the likelihood of injury, while a higher job satisfaction score was linked to a 1.485-fold increase.

Regarding the third question of this study,

Table 15. Pearson correlations between work-related stressors and injury.

	Job Demand	Role Ambiguity	Job Control	Job Certainty	Skill Demand	Social Support	H&D	SCW	ICW	Job Satisfaction
Pearson Correlation	0.229 **	0.069	0.053	0.019	0.009	0.068	0.094	0.024	0.097	0.085
Sig. (2-tailed)	<0.001	0.200	0.324	0.723	0.863	0.207	0.080	0.659	0.070	0.113

** Correlation is significant at the 0.01 level (2-tailed). Note: H&D = Harassment and Discrimination, SCW = Supervisor. Conflicts at Work, and ICW = Interpersonal Conflicts at Work.

presents the Pearson correlation coefficients between self-reported injury and the ten measured work-related stressors. The analysis revealed a statistically significant positive correlation between job demand and injury (r = 0.229, p < 0.001), indicating that higher job demand is moderately associated with an increased likelihood of workplace injury. However, no other stressor demonstrated a statistically significant correlation with injury. Specifically, variables such as role ambiguity (r = 0.069, p = 0.200), job control (r = 0.053, p = 0.324), job certainty (r = 0.019, p = 0.723), skill demand (r = 0.009, p = 0.863), social support (r = 0.068, p = 0.207), health and development (r = 0.094, p = 0.080), safety climate for workers (r = 0.024, p = 0.659), interaction with coworkers (r = 0.097, p = 0.070), and job satisfaction (r = 0.085, p = 0.113) were not significantly correlated with injury. These results suggest that, among the examined stressors, job demand plays the most prominent role in injury occurrence.

5. Discussion

Analysis of the survey data indicates that stress-related factors in the work environment may play a role in predicting injury occurrence among construction workers in Saudi Arabia. In the first model, a statistically significant relationship was identified between work-related stressors and workers’ self-reported injuries. These findings align with earlier research emphasizing the impact of job site demands on worker safety [48], job control [2,9], job certainty, interpersonal conflicts at work, skill demand, and supervisor conflicts at work [49]. On the other hand, in Model 1, there was a low relationship between the average work-related stressor score and the self-reported injury. However, this small effect size suggests that stress is not a direct cause of injury. Instead, it may lead to other symptoms such as fatigue, distraction, or carelessness, which can increase the likelihood of injury [44]. Additionally, this result aligns with earlier research emphasizing the need to examine stressors separately because of their varying levels of impact on safety outcomes [37,44].

Regarding the top key predictive factors of work-related stressors leading to workplace injuries among construction workers in Saudi Arabia, and whether there is a statistically significant relationship between work-related stressors and the self-reported injury rate in Saudi Arabia’s construction industry (i.e., Model 2), the results indicate a statistically significant relationship. Indeed, the logistic regression results (Table 10) revealed that among the examined work-related stressors, job site demand emerged as a statistically significant predictor of self-reported injury (p < 0.001), with an odds ratio of 1.964. This indicates that workers experiencing higher job demands are nearly twice as likely to report injuries. In contrast, other factors such as role ambiguity, job control, and social support did not demonstrate statistically significant associations with injury risk, suggesting a limited predictive value within the context of this model.

While some stressors in the model showed weak or non-significant effects, their presence still contributes to a broader understanding of how various factors relate to injury risk in construction settings. The significant link between job site demands and injury (OR = 1.964, p < 0.001) indicates that heavy workloads and tight deadlines are immediate concerns that can directly influence safety outcomes. Although variables such as job control and social support did not reach statistical significance, this does not mean they are unimportant. Workplace culture, hierarchical structures, or the temporary nature of employment in the Saudi construction industry may shape their limited impact in this context. These findings emphasize the importance of looking at both statistical results and real-world implications when evaluating workplace safety. The substantial predictive value of job site demand suggests it should be a focus of safety improvement efforts, including better planning of work schedules and staffing levels. This study helps highlight which stressors have the most significant impact and points to the need for more detailed or long-term analysis of other psychological factors that may influence worker safety over time.

In this regard, our findings align with those of Leung et al. [48] regarding job site demand. In addition, researchers argue that stress levels related to job demands vary depending on job roles and workplace conditions. Therefore, organizations should implement tailored stress management strategies that address the specific needs of each role, rather than relying on uniform solutions [42]. However, Goldenhar et al. [2] did not find a statistically significant relationship between job site demand and workplace injury. Moreover, Siu et al. [51] asserted that the relationship between job satisfaction and workplace injury is insignificant, which contrasts with the findings of our study. Furthermore, our findings regarding the stressors of job control and job certainty were contrary to those of both Goldenhar et al. [2] and Leung et al. [9], who identified a statistically significant relationship between these two stressors and workplace injuries. On the other hand, our findings align with those of Goldenhar et al. [2] and Leung et al. [9], which suggest that the stressor of social support is not significantly related to workplace injuries.

Moreover, our findings were in agreement with those of Siu et al. [51], who found that the relationship between interpersonal conflicts at work and workplace injury is not significant. In contrast, the findings of Chen et al. [49] suggested a significant relationship between interpersonal conflicts at work and workplace injury. Another point of disagreement was regarding skill demand, where both Chen et al. [49] and Goldenhar et al. [2] asserted a significant relationship between this stressor and workplace injury, in contrast to our findings. Finally, while our findings were in agreement with those of Goldenhar et al. [2] nor Leung et al. [48] that the relationship between the stressor of harassment and discrimination and workplace injury is not statistically significant, our findings differed with previous studies in terms of conflict with supervisor, with Chen et al. [49] identifying a significant relationship between supervisor conflicts at work and workplace injury.

These inconsistencies may be attributed to variations in the demographic characteristics of participants across the different studies. Previous studies have been conducted in various jurisdictions and regions worldwide, including Hong Kong [3,9,48], the Pacific Northwest region of the United States [2], Canada [34,49], and China [51]. In contrast, the present study was conducted in KSA, and as stated earlier, the Saudi construction industry is distinct from other sectors because of the presence of a highly diverse workforce, with individuals from various nationalities. Variations in study settings naturally lead to differences in participant demographics and workplace conditions. Additionally, each cultural context introduces distinct stress factors. Cultural environments differ in physical, economic, and social aspects, as well as in their underlying values and belief systems. These values reflect societal norms about acceptable social and moral conduct. Consequently, since certain values are unique to specific cultures, corresponding behaviors and the work-related stressors associated with them can also be culture-specific. As a result, some differences between cultural contexts can be expected to be reflected in the stress levels, responses, and coping mechanisms observed among individuals [39]. In this regard, Wong et al. [63] argued that cultural backgrounds significantly influence how individuals perceive and respond to stress, leading construction site workers to adopt varying coping strategies based on these differences.

Given that migrant workers are prevalent in the construction sector in Saudi Arabia, their perceptions and stress management mechanisms are likely to be influenced by diverse cultural factors at the employee, project, and organizational levels [26,27,64]. Meanwhile, Loosemore et al. [65] have noted that workers in workplaces with a high degree of cultural diversity tend to be more supportive and protective of coworkers of a similar ethnic or cultural background. There is substantial evidence of inequity in the workplace based on demographic factors, whether in the form of unfair pay, inequity in terms of upward mobility within an organization, or offensive graffiti and racist jokes [65,66]. Language barriers have also been identified as a significant challenge in forging social connections between different cultural groups, and they have also been shown to pose a direct safety hazard in the workplace [65,66,67,68]. In this context, it is worth noting that Saudi Arabia’s construction industry employs about 40% of the working population, making it one of the leading industries in Saudi Arabia in terms of its share of the overall workforce [26]. Around 88% of these workers are foreign nationals [26]. Given the high proportion of foreign labor in Saudi Arabia’s construction sector and the industry’s considerable cultural diversity, the presence of these stressors among construction workers in the Kingdom is to be expected.

This study contributes to filling a critical gap by developing and validating a work-related stressor scale tailored to Saudi Arabia’s construction industry. From a research perspective, the use of logistic regression allowed for the identification of the unique impact of each stressor while controlling the influence of others. The consistent significance of job site demand, even after accounting for other variables, highlights its strength as a predictor of injury risk. However, some stressors may influence injury outcomes indirectly, for instance, through changes in safety behaviors. These potential indirect or interactive effects were not captured in the current model and could be more effectively explored in future studies using advanced statistical techniques and predictive modeling approaches. The insights gained can inform strategic interventions by construction firms and regulatory authorities, particularly the Ministry of Human Resources and Social Development, aimed at reducing stress-related injury risks. The study’s findings have important implications for construction safety, emphasizing the need to support workers’ psychological well-being alongside traditional safety initiatives. In the context of construction safety in Saudi Arabia, the findings underscore the critical need to address work-related stressors broadly, with particular emphasis on the following areas:

• Mitigating Job Site Demands: Given the strong correlation between job site demand and injury, firms must ensure adequate workforce allocation and realistic deadlines. Excessive working hours should be curtailed to prevent fatigue-related accidents. Employers must also address environmental stressors, especially extreme heat, by providing cooling solutions and enforcing midday work restrictions.
• Improving Job Certainty: Short-term and daily employment contracts create insecurity that exacerbates stress. Employers should aim to provide long-term, stable employment options, which could alleviate psychological burdens and reduce injury risk.
• Enhancing Worker Motivation and Satisfaction: Fair wages, opportunities for professional development, proper facilities, and competent supervision contribute significantly to job satisfaction. These factors not only improve productivity but also play a role in safety, as engaged and motivated workers are more likely to comply with safety protocols.
• Promoting a Supportive Work Environment: Strategies to enhance communication across language barriers, such as multilingual signage or interpreter support, can improve both morale and safety outcomes. Cultural sensitivity training may also help reduce interpersonal conflict and foster inclusion.
• Integrating Mental Health into Safety Programs: Construction safety initiatives must go beyond physical safety measures to include mental health support. As prior studies suggest, companies with strong safety climates that incorporate psychological wellness programs tend to perform better overall [34].

Enhancing worker satisfaction in these areas not only encourages optimal performance but also helps reduce the risk of workplace accidents.

6. Conclusions

Construction work is inherently high-risk because of its physically demanding tasks, extended working hours, and constant exposure to hazardous conditions. These challenges are compounded by additional stressors, including excessive workload, poor physical environments, interpersonal conflicts, and inadequate access to safety equipment. This study builds on the existing literature to identify and conceptualize ten prevalent work-related stressors commonly encountered in the construction sector: job site demand, role ambiguity, job control, job certainty, skill demand, social support, harassment and discrimination, conflicts with supervisors, interpersonal conflicts, and job satisfaction.

Focusing on the construction industry in Saudi Arabia, this study aimed to assess the impact of these stressors on worker safety and to determine which stressors most strongly predict workplace injury. To address these objectives, two guiding research questions were developed: (1) Is there a statistically significant relationship between work-related stressors and the frequency of self-reported injuries among construction workers in Saudi Arabia? and (2) Which stressors serve as the most influential predictors of self-reported injury? A structured, self-administered survey was distributed in person to 349 randomly selected construction workers across 16 worksites in Saudi Arabia. The survey instrument included validated scales measuring each of the ten stressor domains.

The collected data were analyzed using logistic regression to evaluate the influence of work-related stressors on construction safety. All responses were measured using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). At the same time, Pearson correlation analysis was employed to assess the relationships between stressors and the frequency of self-reported injuries. While we acknowledge that Likert-scale data are ordinal and that the injury variable is binary, which can raise concerns about the appropriate correlation method, the Pearson correlation is commonly used in research involving Likert scales treated as interval data, especially when the scale has five or more points, as in our survey. Pearson correlation provides a straightforward measure of linear association that is interpretable and comparable with the prior literature in construction safety research. Additionally, the Likert items were drawn from well-established, validated scales discussed in the literature. Given that the primary purpose was to explore relationships rather than infer causality or complex modeling. Given the binary nature of the variables involved, both research questions were analyzed using IBM SPSS Statistics. The findings indicate that work-related stressors can serve as predictors of workplace injury among construction workers in Saudi Arabia. Model 1 revealed a statistically significant relationship between stressors and self-reported injury, though the overall effect size was modest. Model 2, developed to address the second research question, was found to be statistically significant, identifying job site demand and job satisfaction as the most influential stressors among the ten examined. Both factors were positively associated with an increased risk of workplace injury.

Despite these insights, several limitations must be acknowledged. First, this study relied on self-reported data, which may be subject to bias or inaccurate recall. Second, the cross-sectional nature of this study limits the ability to draw causal inferences. Third, although the sample included a variety of nationalities, this study did not account for cultural or language differences that may influence stress perception and reporting behavior. Lastly, while the models were statistically significant, the observed effect sizes were relatively small, suggesting that additional unmeasured factors may also contribute to workplace injuries.

Future research should build on these findings by employing longitudinal designs to explore how chronic exposure to stressors influences injury risk over time. Mixed-method approaches incorporating qualitative interviews could offer more profound insights into how workers experience and cope with stress. Future studies should also investigate how nationality, language proficiency, and company safety policies interact with stress and injury outcomes, particularly given the cultural diversity of Saudi Arabia’s construction workforce. Finally, policy-focused research can help evaluate the effectiveness of targeted interventions aimed at mitigating key stressors, such as workload management and job satisfaction enhancement.

By identifying the stressors most relevant to injury risk, this study offers practical guidance for employers, regulators, and policymakers seeking to improve occupational safety in Saudi Arabia’s construction sector. Prioritizing reductions in job site demand and fostering greater job satisfaction may not only enhance worker well-being but also reduce the burden of occupational injuries across the industry.