The Influence of Nudges on Workers’ Safety Behavior in the Construction Industry ^†

Abstract

The nudge theory has been applied to improve safety behavior in various industries. We implemented the nudge theory in the construction industry to improve worker safety. Nudges for construction safety were grouped into three categories and used in a project. The application of nudges improved workers’ safety behavior, particularly in highlighting control measures. The background analysis results reveal that the implementation of nudges did not vary by gender, nationality, or age, demonstrating its high universality. The results of this study offer a reference for stakeholders in the construction industry who are aiming to bolster worker safety.

1. Introduction

Construction engineering, characterized by labor-intensive processes, diverse machinery, and complex interfaces, is considered a high-risk workplace. According to relevant studies in the literature [1,2], unsafe behavior is the primary cause of accidents in the construction industry. Therefore, it is crucial to enhance workers’ safety behavior and reduce occupational accidents. There are laws, regulations, standards, and other mandates to secure workers’ safety behavior, but these mandates are top-down commands. To encourage workers to voluntarily change their behaviors, the nudge theory can be used as an emerging and effective method [3].

The nudge theory, rooted in behavioral economics and psychology, employs ingenious design to assist individuals in making correct decisions. The nudge theory has been applied to improve safety behavior in various industries. Although its application in the construction industry is in its infancy, it has already shown impressive results. However, nudge measures must possess practical and easily promotable characteristics to effectively change safety behavior [4].

Therefore, we categorized construction safety nudges into Alerting Potential Hazards, Highlighting Control Measures, And Promoting Health Care and proposed practical nudges through collaborative discussions with construction safety experts. We established the nudge measure scale and the worker safety behavior scale and used them in a project to verify the effect of using nudges. Recommendations were made for strategies to implement the nudge theory based on research results.

2. Literature Review

2.1. Nudge Theory

The nudge theory was proposed by Thaler and Sunstein [5], originating from behavioral economics and psychology. Through ingenious design, it assists people in making robust decisions. The UK announced the establishment of the ‘Nudge Unit’, composed of scientists, psychologists, and policy experts, to apply the nudge theory to government policy implementation to enhance execution capability and service efficiency and yielded high cost-effectiveness for the UK.

Improving workers’ safety behavior is an essential research topic regarding the nudge theory. Prabowo and Syaifullah [3] believe that unsafe behavior is the most critical factor leading to occupational accidents, and nudges are emerging and effective in eliminating or reducing unsafe behavior. Lindhout and Reniers [4] regard the nudge theory as an effective safety management tool capable of influencing workers’ behavior and compliance. They organized nine design principles of nudges based on the literature review and believed that more practical and easily promotable nudges effectively change safety behavior. Szaszi et al. [6] investigated the effectiveness of the nudge theory as a safety behavior intervention measure. They found that the nudge theory, as an intervention, is low-cost and easy to implement and effectively influences the target group to unconsciously perform specific behaviors.

The nudge theory has been applied to improve safety behavior in various industries. Although its application in the construction industry has just begun, it has already shown astonishing results. The Royal Dutch Shell’s Southbank office refurbishment project is a well-known case. The results of the nudges were clear: the levels of unsafe behaviors while working at height were reduced by 82% and by 93% for material movements in just 12 weeks [7]. The project also won the International Safety Award with Merit from the British Safety Council and the Health, Safety & Wellbeing Initiative of the Year award at the Association for Project Safety’s National CDM Awards 2018.

Prabowo and Syaifullah [3] believe that nudges are effective in eliminating or reducing unsafe behavior in the construction industry. They explored the example of increasing the usage of personal protective equipment (PPE). They posted workers’ hearing test results with and without PPE on the bulletin board as nudges. The pre-test and post-test comparisons showed that the nudges were gentle and effective interventions to eliminate workers’ unsafe behavior. They also believed that nudges must be integrated into risk management and occupational safety and health management systems to improve organizational safety culture. Talabi et al. [8] believe that safety design and nudges must be implemented together, with safety design being used to eliminate construction risks while promoting nudges to reduce workers’ unsafe behavior and reduce occupational accidents. Talabi et al. [9] considered nudges as effective safety interventions to enhance workers’ safety behavior; they claimed that nudges must not be regarded directly as control measures but as reinforcements of control measures. Digital technology is gradually becoming a medium for nudges. Nioi et al. [10] used a mobile app to push information, such as UV radiation, sun safety, dietary health, and daily UV index, in the workers’ area as nudges.

The nudge theory has been successfully applied to improve construction workers’ safety behavior. Through development, digital technology has also become an essential medium for nudges. Moreover, the literature shows that practical and feasible nudging measures effectively changed safety behavior.

2.2. Safety Behavior

Many studies demonstrated that improved workers’ safety behavior effectively reduced the occurrence of occupational accidents in construction. In ref. [11], safety behavior activities were implemented in two construction sites, each with approximately 200 workers, mainly from Bangladesh, China, Singapore, and Thailand. An online safety behavior tracking system was established to support the activities. The results show that through safety behavior activities and online tracking systems, continuous improvement in construction site safety was achieved. In a large construction site in Iran, the effectiveness of safety behavior was verified [12]. The results show that the safety index of the experimental group was significantly higher than that of the control group. Safety behavior activities at two construction sites in Singapore showed that the overall safety scores of the two projects were improved after implementing safety behavior activities [13]. In a construction site in Hong Kong, the impact of managerial behavior on workers’ safety behavior was investigated [14]. The results show that managerial behavior affected the effectiveness and sustainability of workers’ safety behavior, thereby improving the safety performance of the construction industry.

In addition to the effectiveness of safety behavior, the indicators of workers’ safety behavior in the construction industry are also important. Chouudhry proposed using PPE; housekeeping; access to heights, plant, and equipment; and scaffold-type worker safety behavior indicators as interventions [15]. With these interventions, safety behavior significantly increased, and frontline worker safety was improved. Li et al. [1] proposed four categories and 10 unsafe behavior indicators, including falls, collisions, traffic accidents, and a lack of sensor detection. They analyzed the changes in indicator data to evaluate the safety performance of construction projects. The research results show that the safety index improved after promoting safety behavior activities, and this index is used in the construction industry. Guo et al. [16] established effective safety behavior indicators for Singapore tunnel engineering, including lifting operations, excavation, high-altitude operations, work platforms and passages, handling, hot work, factories and equipment, traffic management, personal protective equipment, and nine other categories. Ting et al. [17], based on previous studies, made adjustments according to actual application feedback, proposing four categories of safety behavior indicators: working environment, tools and equipment, body movement, and personal protective equipment. These indicators were applied to 19 engineering projects for over 10 years. The overall Total Recordable Case Rate (TRCR) significantly decreased after promoting safety behavior. The safety behavior indicators proposed by Ting et al. showed reliability and validity [18,19].

3. Methodology

3.1. Research Objective

We proposed practical construction safety nudges and investigated whether these nudges affected workers’ safety behavior. We also explored whether different background conditions, such as gender, nationality, age, and position, result in variations in safety behavior performance.

3.2. Nudges for Construction Safety

Practicality and ease of promotion are key success factors for nudges [4]. Therefore, through collective discussions among safety experts, we categorized construction safety nudges into three classes: Alerting Potential Hazards, Highlighting Control Measures, and Promoting Health Care. To date, many practical nudges have been proposed (

Table 1. Construction safety nudges.

Alerting Potential Hazards: Setting up warning slogans, signs, signals, etc., to alert workers to potential hazards.
Paint safety passage rebar to prevent falls.	Add glow stickers by slogans for visibility.	Use sensor alerts for hazards.
Mark pier levels against drowning.	Install lines at platform road for alerts.	Add cameras in blind spots for collision warnings.
Highlighting Control Measures: Enhance controls with added design for effectiveness.
Use mesh on bridges for safety.	Lay mesh on rebar to prevent falls.	Paint belt hooks for compliance checks.
Paint crane supports for collapse prevention.	Install LED lights for nighttime construction safety.	Post warning lights on electric boxes to prevent shock.
Promoting Health Care: Various nudges concerned with the physical and mental health of workers.
Use smart mist system for thermal stress prevention.	Give supervisors heat index detectors for break reminders.	Post urine charts in restrooms for hydration reminders.
Color-code helmets for worker types.	Install fans for heat safety.	Set up solar rest area for workers.

).

3.3. Nudge Measure Scale and Worker Safety Behavior Scale

The nudge measure scale and the worker safety behavior scale were designed using a five-point scale, where 1 represents ‘strongly disagree’, 2 represents ‘disagree’, 3 represents ‘neutral’, 4 represents ‘agree’, and 5 represents ‘strongly agree’. After a discussion with experts, the nudge measure scale was defined as shown in

Table 2. Nudge measure scale.

X1 Alerting Potential Hazards

X1.1 Visibility of the location of the hazard

X1.2 Awareness of the type of hazard

X1.3 Understanding of the potential impact of the hazard

X2 Highlighting Control Measures

X2.1 Awareness of control measures

X2.2 Recognition of the key protection aspects

X2.3 Understanding of the usage of control measures

X3 Promoting Health Care

X3.1 Awareness of health care measures

X3.2 Understanding of the purpose of health care measures

X3.3 Ability to remind oneself to pay attention to health status

, including the labels X1 (Alerting Potential Hazards), X2 (Highlighting Control Measures), and X3 (Promoting Health Care).

The worker safety behavior scale demonstrated practicality and effectiveness in [17]. Therefore, we adjusted the worker safety behavior scale (

Table 3. Worker safety behavior scale.

Y1 Working Environment

Y1.1 Notice the Condition of the Surfaces

Y1.2 Notice Extreme Temperature/Stress Exposure

Y1.3 Ensure Appropriate Lighting

Y1.4 Observe Signs and Barriers

Y1.5 Store Materials in Order

Y1.6 Segregate Garbage or Waste in Place

Y2 Tools and Equipment

Y2.1 Select the Correct Tools/Equipment

Y2.2 Inspect the Conditions of Tools/Equipment Before Use

Y2.3 Use Tools/Equipment Correctly

Y2.4 Operate Machines/Vehicles Well

Y3 Activity Procedure

Y3.1 Avoid Dangerous Areas

Y3.2 Stay Away from Hot Work Fire

Y3.3 Avoid Taking Shortcuts

Y3.4 Proceed and Operate with Care

Y3.5 Be Aware of Moving/Rotating Machines

Y3.6 Notice Proper Body Posture

Y3.7 Use Ladders Properly

Y3.8 Clime with Right Body Movement

Y3.9 Communicate Proactively

Y4 Personal Protective Equipment

Y4.1 Select the Correct PPE for the Job

Y4.2 Inspect the Condition of PPE Before Use

Y4.3 Use PPE Correctly

), using labels including Y1 (working environment), Y2 (tools and equipment), Y3 (activity procedure), and Y4 (personal protective equipment).

3.4. Statistical Analysis

We conducted descriptive statistics to illustrate the background information of the respondents and used t-tests or an ANOVA to examine differences in worker safety behavior scores based on different backgrounds (such as gender, nationality, age, and position). A correlation coefficient analysis was used for nudges (X1 to X3) to test multicollinearity issues. Subsequently, a regression analysis was conducted to explore the relationship between nudges and workers’ safety behavior (Y1 to Y4), examining their explanatory power and statistical significance. Finally, a correlation coefficient analysis was also applied to workers’ safety behavior to test its correlation.

4. Results and Discussions

4.1. Background Analysis

We selected a project located in Tainan to investigate the changes in workers’ safety behavior after the implementation of nudges. A total of 81 valid questionnaires were collected. The demographics of the respondents were as follows:

• Seventy-nine males and two females;
• Eleven Taiwanese people and seventy foreign nationals;
• Four respondents aged 20–30 years old, eighteen respondents aged 30–40 years old, forty-five respondents aged 40–50 years old, twelve respondents aged 50–60 years old, and two people over 60 years old;
• Seventy-four workers and seven managers.

Although most of the questionnaires were answered by workers regarding the impact of the nudges on their safety behavior, several were filled out by managers, reflecting their observations on whether the workers’ safety behavior was affected.

4.2. Workers’ Safety Behavior

The average value of workers’ safety behavior (Y) was 4.56, with the average value for working environment (Y1) being 4.55, the average value for tools and equipment (Y2) being 4.57, the average value for activity procedure (Y3) being 4.57, and the average value for personal protective equipment (Y4) being 4.56. These indicated that the implementation of nudges on the construction site enhanced workers’ safety behavior. The top four highest-scoring items included Use PPE Correctly (Y4.3), Proceed and Operate with Care (Y3.4), Store Materials in Order (Y1.5), and Operate Machines/Vehicles Well (Y2.4). For example, adding conspicuous markings to personal protective equipment (such as spray painting safety belt hooks) as a nudge guided workers to unconsciously use PPE correctly, as shown in Figure 1.

In workers’ safety behavior, the Pearson correlation coefficients for Y1 to Y4 were calculated, and a Bonferroni correction with a threshold was set at 0.05. The results revealed a significant correlation between working environment (Y1) and activity procedures (Y3) (r = 0.50), and also between tools and equipment (Y2) and activity procedures (Y3) (r = 0.39). Figure 2a illustrates the establishment of a cross-bridge passage using a diamond lattice grid as a nudge. Such a nudge affected the Notice the Condition of the Surfaces (Y1.1) and Avoid Taking Shortcuts (Y3.3) items. Figure 2b shows the posting of an electrical map on a temporary electrical box as a nudge. This nudge influenced the Inspect the Conditions of Tools/Equipment Before Use (Y2.2) and Proceed and Operate with Care (Y3.4) items.

T-tests or an ANOVA were used to test workers’ safety behavior based on different backgrounds (gender, nationality, age, and position). The results indicate that only different positions exhibited significant differences, while other backgrounds showed no significant differences in the performance of workers’ safety behavior. The nudges had universality and can be applied to most construction sites. In terms of different positions, the worker safety behavior scores determined by on-site managers (4.73) were higher than the self-evaluated scores by the workers (4.55). This result may be due to the on-site managers having a better understanding of the purpose and usage of the nudges, resulting in higher observed scores. Therefore, it is necessary to enhance the promotion of various nudges to workers to improve their safety behavior.

4.3. Regression Analysis

A correlation coefficient analysis was conducted for the nudges (X1, X2, and X3). The correlation coefficient between X1 and X2 was 0.327, that between X1 and X3 was 0.387, and that between X2 and X3 was 0.372. The correlations among the variables were not high, indicating that the issue of multicollinearity among the independent variables was not severe, and the regression analysis results could be relatively trusted. The regression analysis result regarding the nudges (X1, X2, X3) and workers’ safety behavior (Y) is shown in Equation (1).

$$Y=1.8374+0.2111X1+0.2193X2+0.1688X3$$

(1)

The coefficient of X1 is 0.2193, with a p-value of less than 0.001, indicating a statistically significant positive correlation between X1 and Y, and the same applies to X2 and X3. The R-squared value is 0.591, demonstrating reasonable explanatory power. According to Equation (1), the influence of nudges was in the order of Highlighting Control Measures (X2), Alerting Potential Hazards (X1), and Promoting Health Care (X3). Thus, construction sites may prioritize nudges that highlight control measures as they have the greatest impact on workers’ safety behavior.

The statistical analysis results are summarized in Figure 3, showing that the implementation of nudges can indeed enhance workers’ safety behavior, with the Highlighting Control Measures (X2) item having the greatest impact. The most effective items included Use PPE Correctly (Y4.3), Proceed and Operate with Care (Y3.4), Store Materials in Order (Y1.5), and Operate Machines/Vehicles Well (Y2.4).

5. Conclusions

Based on a literature review, we posited that construction safety nudges enhanced workers’ safety behavior, thereby reducing the occurrence of occupational accidents. Therefore, through expert discussions, the nudges were categorized into three types, namely Alerting Potential Hazards, Highlighting Control Measures, and Promoting Health Care, and many practical nudges were proposed. Additionally, we developed scales for nudge measures and workers’ safety behavior as measurement tools. The results from the investigation and analysis indicate that nudges improved workers’ safety behavior, with the Highlighting Control Measures being the most influential nudge. Construction sites may prioritize implementation from this category. The nudge measures proposed in this study are characterized by their practicality and high cost-effectiveness. They were given feedback from the construction industry for universality. Although we focused on a single construction site, construction safety nudges need to be promoted nationwide. More extensive survey data are needed to validate the research hypotheses. Once a large amount of data are collected, structural equation modeling can be employed for further analysis.