You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Buildings
Download PDF
  • Article
  • Open Access

Published: 22 April 2025

The Impact of Work Sequence-Based Safety Training on Workers’ Cognitive Effectiveness at Construction Sites

,
,
,
and
1
Department of Safety Engineering, Seoul National University of Science and Technology, Seoul 01811, Republic of Korea
2
Department of Civil Engineering, Kangwon National University, Samcheok 25913, Republic of Korea
3
Samsung C&T Corporation, Tower B, 26, Sangil-ro 6-gil, Gangdong-gu, Seoul 05288, Republic of Korea
*
Authors to whom correspondence should be addressed.
Buildings2025, 15(9), 1409;https://doi.org/10.3390/buildings15091409
This article belongs to the Section Construction Management, and Computers & Digitization
Version Notes
Order Reprints

Abstract

Daily pre-work safety training for workers is essential due to their exposure to various hazardous factors. However, existing training methods often fail to effectively deliver information in a way that supports workers’ learning and comprehension, resulting in poor job performance and a lack of adherence to safety protocols. This study explores a novel approach to structuring safety training by examining the cognitive characteristics of workers in relation to the placement order and composition of training materials. An educational recall experiment involving 660 construction workers was conducted to evaluate the impact of training material design on content retention. The results showed that the average number of recalls was significantly higher when the training content was structured according to the actual work sequence, compared to content organized in a random order. Additionally, separating execution standards from safety standards led to better recall than when the two were integrated. These findings highlight the importance of aligning training materials with workers’ cognitive processing patterns. This study contributes to both theory and practice by providing evidence-based guidance for improving pre-work safety training. The proposed method offers a practical solution for small- and medium-sized construction companies that often lack the resources to develop effective training materials.

1. Introduction

The construction industry plays a crucial role in global economic growth but is also recognized as one of the most hazardous industries [1]. As construction sites become larger and more complex, workers are exposed to various risks, making safety concerns and skilled labor shortages critical issues [2,3,4,5]. Due to the nature of construction projects, where teams perform different tasks and team members change throughout the process, strict adherence to work procedures and safety standards is essential [6].
The construction industries in the United States, Europe, Denmark, and South Korea have higher industrial accident rates than other industries [7,8,9,10,11]. The primary causes of these accidents are hazardous working environments and unsafe worker behavior [12]. Many construction workers engage in unsafe practices due to a lack of risk awareness, while factors such as alcohol consumption, language barriers, and low education levels further increase accident frequency and severity [13,14,15,16,17,18]. In South Korea, major causes of fatal construction accidents include defective safety protection measures and structural deficiencies, as well as unsafe behaviors such as ignoring hazards or improperly using protective equipment [11]. Studies indicate that 88% of construction site accidents and fatalities are caused by unsafe behaviors, with over one-third of workers engaging in risky work practices [19,20,21,22,23,24,25,26].
To address these issues, various approaches have been explored to reduce unsafe behaviors. Behavior-Based Safety (BBS) programs have proven effective in promoting and maintaining safe work practices [27,28,29,30]. Additionally, incentives and safety training play a significant role in influencing worker behavior [31,32,33]. However, BBS research does not fully explain the underlying causes of unsafe behaviors or the personal factors linked to safety culture [34,35]. As a result, Cognition-Based Safety Management has gained attention for its ability to analyze workers’ cognitive processes and decision-making mechanisms, helping identify the fundamental causes of unsafe behavior [5,25,34,35,36,37].
Effective safety training is essential to addressing these concerns [38]. A structured safety training program not only enhances workers’ safety knowledge and behavior, improving overall safety performance, but also boosts confidence and encourages active participation in safety initiatives [39,40,41]. Moreover, regular safety training helps establish clear safety guidelines and fosters a stronger safety culture through active worker engagement [42,43,44].
Despite these efforts, existing safety training programs often suffer from low worker engagement, a formalized approach, and inefficiencies in training methods [45]. Although safety training is widely implemented in construction sites [46], 45% of fatal construction accidents in 2021 were attributed to inadequate inspections, poor maintenance, and insufficient preparation—all factors directly related to training deficiencies [11]. This highlights the need for more effective safety training approaches.
Tool Box Meetings (TBMs) serve as daily pre-task safety training sessions, ensuring that workers understand work procedures and safety standards [47]. TBM can enhance team communication, improve risk awareness, and help prevent industrial accidents. Moreover, using pre-prepared, site-specific training materials can increase worker engagement [48]. Notably, an interactive TBM approach that encourages worker participation and feedback has been found to have an immediate positive effect on worker behavior.
However, TBM practices currently face several challenges. Most team leaders conduct TBM verbally based on personal experience, with minimal effort in creating and distributing systematic training materials. Since TBM takes place just before work begins, training materials must be designed for quick comprehension of work procedures and safety standards. There is a lack of standardized training materials tailored to specific tasks, leading to the omission of crucial safety elements [49]. Therefore, research has been conducted on TBM in relation to safety training content [50,51].
To improve TBM effectiveness, structured and standardized training materials should be developed, incorporating learning theories to enhance comprehension and memory retention. However, research on the effectiveness and delivery methods of TBM training remains insufficient.
This study aims to develop TBM training materials based on learning theories and analyze their impact on workers’ memory retention after training. Ultimately, this research seeks to enhance TBM training effectiveness, maximize safety education outcomes, improve workers’ safety performance, and reduce industrial accidents.

2. Literature Review

2.1. Pre-Task Construction and Safety Training Requirements

According to the Construction Technology Promotion Act, pre-construction and safety training must be conducted before commencing construction work on the same day. The training should include an understanding of the construction methods to be used that day, detailed procedures based on construction drawings, and important technical precautions. The training must be conducted either by preparing educational materials or verbally by the construction supervisor (safety officer under the Construction Technology Promotion Act) [52].
TBM is conducted as a pre-task briefing in most countries, typically lasting 10 to 15 min. During this session, workers are educated on the day’s work procedures, past accident cases, and important safety precautions. Therefore, given the short duration of TBM, it is crucial for workers to retain and learn the training content effectively [47,48,49,50,51].

2.2. Cognitive-Based Safety and Learning Theory

Learning is portrayed as a complex process by which environmental stimuli are apprehended through successive stages of information processing [53], enabling individuals or organizations to interact with their environment, acquire information, and, based on this foundation, cultivate new behaviors or constructs. This progression involves key processes such as memory, attention, generalization, experience, transfer, and observation [54].

2.2.1. Learning Theory in Cognitivism

In cognitivism, learning is envisioned as a transformation of knowledge achieved through the processing of information, distinct from the behaviorist view that learning outcomes stem from the pairing of stimuli and responses. Cognitivism asserts that learning involves the alteration of the learner’s internal cognitive structures, thereby framing the learning process within the bounds of information processing [55]. This perspective emphasizes the importance of the learner’s cognitive processes, such as thought, memory, and information handling, positioning the individual’s cognitive activities as central to the learning experience. Cognitive learning theories conceive of learners as active participants, who not only actively receive new information but also reorganize and reinterpret this information in a manner that is unique to their personal understanding [56].
Core to the cognitive learning theory are the concepts of schema, the information processing model, and learning strategies. Schemas are understood as cognitive frameworks pertaining to objects, situations, people, and their interrelations, acting as interpretive structures through which individuals assimilate or reject new information [57]. Learners assimilate new information by connecting it with existing schemas, thus ascribing meaning and facilitating comprehension. The information processing model delineates cognitive processes as a sequence from attention to perception and rehearsal, then encoding, and finally retrieval. Learning strategies encompass the cognitive tasks of planning, organizing, and monitoring learning activities to achieve educational goals, playing a vital role in enhancing the efficiency and effectiveness of learning.
Learning theory in cognitivism is explained by theories such as Gestalt psychology, information processing theory, metacognition, and cognitive load theory.

2.2.2. Gestalt Psychology

When observing objects, humans exhibit a tendency to organize incoming external stimuli into coherent forms based on perceptual structuring. For shapes perceived by our eyes due to external stimuli to be recognized as a single pattern, the characteristics of the elements that compose the pattern must be interrelated and organized. This process and tendency towards recognizing and perceiving a single form is referred to as grouping. The principles of grouping—proximity, similarity, closure, and continuation—facilitate this process [58,59].
From a pedagogical perspective, grouping learning materials can make the organization of educational content psychologically significant to learners by employing concepts of proximity, similarity, and continuation. The human information processing system, upon encountering learning opportunities, involves environmental stimulus-triggering information that is then stored in memory, processed, and later retrieved for recall. This parallels the manner in which information is processed in the brain, akin to computational operations. Cognitive Information Processing Theory centers on how attention is directed towards surrounding events, how learned content is encoded and associated with existing knowledge in memory, how new information is stored, and the mechanisms of retrieval [60]. Therefore, it was determined that safety training would have an impact on task organization.

2.2.3. Information Processing Model and Metacognition

The information processing model, a representative model of cognitive learning theory, is a model that represents a conceptual structure for explaining how information is processed and interpreted in the human brain. It comprises three key components: storage, cognitive processes, and metacognition [61].
In storage, sensory memory retains sensory information for a very brief period, thereby extending the time necessary for pattern recognition. Here, ‘recognition’ entails classifying a stimulus into various forms based on its appearance. If information stored in sensory memory is not recognized during the pattern recognition stage, that sensory information vanishes [62]. In the cognitive process, the attention process influences the selection of information, rehearsal impacts the degree of internal connections formed with new information, and the encoding process affects the integration of new information with pre-existing knowledge [63].
Metacognition refers to an individual’s ability to organize their cognitive structure, facilitating a strategic review of their learning process [64,65]. Cognitive load denotes the total mental effort required by humans to process information, necessitating the learner’s working memory for information to be processed. When the amount of information that needs to be processed exceeds the capacity of working memory, cognitive load occurs [66]. Exceeding the limits of cognitive load leads to cognitive overload, a primary cause of reduced learning effectiveness. The aim is to eliminate cognitive overload to enable effective learning and the design of optimal learning strategies [67].

2.2.4. Cognitive Load Theory

Cognitive load theory posits that while working memory has limited capacity, long-term memory’s capacity is limitless. Learners transfer information processed in working memory to long-term memory, thereby gaining concepts. Acquiring these concepts allows learners to process information through long-term memory, thus expanding the capacity of working memory. Consequently, the cognitive load experienced by learners decreases, leading to quicker information processing and enhanced cognitive abilities [68].
Cognitive load is categorized into intrinsic cognitive load, extraneous cognitive load, and germane cognitive load. The cumulative cognitive load from these three types cannot exceed the total cognitive capacity [66]. Intrinsic cognitive load is associated with the complexity inherent in a task or the characteristics of the learning content, typically arising from the difficulty of the task and the amount and interactivity of the information comprising the learning content. Extraneous cognitive load varies with the learning method and is induced by poor instructional design; thus, it can be improved through enhancements in teaching methods, presentation of learning content, and learning strategies. Finally, germane cognitive load directly relates to the effort expended by learners to understand the learning content, signifying the intellectual effort invested for comprehension. Given the limited capacity of working memory, instructional designs should aim to reduce extraneous cognitive load, enabling learners to induce germane cognitive load [69].

2.2.5. Forgetting

Forgetting related to recall is considered in learning contexts [70,71] and has been addressed in sectors such as construction [67], the nuclear industry [72], and healthcare [73]. Forgetting can be delayed through regular and repetitive training [74,75].

3. Scope and Design of Experiments

3.1. Research Issue

The current safety training conducted for workers at construction sites is primarily aligned with the requirements for training duration and content as stipulated by the Construction Technology Promotion Act [76] and the Occupational Safety and Health Act [77]. However, there is a notable deficiency in providing systematic and efficient training methodologies prior to work, resulting in a lack of effectiveness in workers’ cognition and learning.
The daily pre-work training content for construction site workers typically encompasses an understanding of the day’s work methods, attention to technical precautions in construction, and primarily focuses on safety standards in TBM to ensure safety. This study categorizes the training content delivered to workers into execution standards and safety standards, defined as follows:
  • Execution Standards: technical guidelines and criteria that must be adhered to ensure the stability and quality of construction works.
  • Safety Standards: guidelines and standards that workers must comply with to maintain a safe and comfortable work environment and to protect life.
Given the practical challenges of separately conducting these two training sessions before daily work on construction sites, TBMs are predominantly used to convey both types of training content. Excluding time allocated for preparation and warm-up, the actual training duration within a TBM ranges from 5 to 10 min. This research aims to explore the impact of the composition order of training materials in TBMs on construction sites, based on the characteristics of learning theories, on the effectiveness of worker education. Furthermore, it seeks strategies to enhance the learning outcomes (improvement of memory activities) for workers in construction site safety training.

3.2. Memory Recall Experiment

Figure 1 illustrates the recall experiment and response analysis process. Participants were assigned to four groups, each receiving different training materials. After instruction, participants completed response sheets. The collected data were analyzed based on demographic characteristics and recall performance, using statistical methods such as t-tests and one-way ANOVA.
Figure 1. Experiment flow chart.

3.2.1. Experiment Overview

This experiment aimed to assess whether the content related to execution standards and safety standards taught to workers during TBMs is effectively delivered and learned. The experiment hypothesized various delivery methods based on the composition of training materials and conducted recall experiments. The results of these recall experiments were analyzed through the count of recalls (recall rate) related to the training content. The research hypotheses were established as follows and were tested through the experiment:
  • Safety training materials organized according to the work sequence will exhibit a higher recall rate post-training than those not organized in such a manner.
  • Training materials that are separately arranged for execution standards and safety standards will demonstrate a higher recall rate post-training than materials that are not separated.
  • Differences in recall rates for training materials will arise based on individual characteristics.
To verify these hypotheses, experimental groups were formed to analyze differences in recall information among them. After categorizing groups based on the delivery method according to the work sequence and a random organization of training materials, new workers were trained, and the cognition of the educational content was verified to ascertain the level of cognition and the correlation among components.

3.2.2. Experiment Material

To determine the impact of cognitivist learning during the process of workers being educated in TBMs, an experimental comparison was conducted based on training materials reflecting characteristics such as work sequence and clustering. The materials were structured into four types, incorporating the findings that suggest a meaningful information (chunk) limit in short-term memory is 7 ± 2 items (5–9 items) [78,79]. Accordingly, the training materials, including execution standards and safety standards, were standardized to have eight slides for the experiment.
The training materials selected for the experiment were focused on the process of steel structure installation currently being carried out at the experiment site. The installation sequence was chosen to include eight stages, site delivery/material transportation, anchor bolt installation, preparation and verification of installation materials, column erection, assembly of large/small beams, main bolting operations, welding tasks, and deck plate work. The training material for each stage was composed of four items which included two work rules and two safety rules. Figure 2 illustrates a portion of the training materials.
  • Type A: structured based on the work sequence, integrating two execution standards and two safety standards for each stage.
  • Type B: structured based on the work sequence, separating two execution standards and two safety standards for each stage.
  • Type C: arranged in a random order irrespective of the work sequence, integrating two execution standards and two safety standards per chart.
  • Type D: arranged in a random order irrespective of the work sequence, separating two execution standards and two safety standards per chart.
Figure 2. Training materials.
Type A training materials were structured to integrate two execution standards and two safety standards per construction sequence stage. In contrast, Type B materials, while maintaining the same stage-by-stage structure as Type A, differed by explaining all execution standards first, followed by detailing the safety standards relevant to each construction stage. Type C and type D training materials were organized randomly without considering the work sequence. However, to maintain uniformity across all experimental groups, the materials were composed of eight charts (four standards per chart), with the content equivalent to type A and type B. For Type C, regardless of the work sequence, each training chart was structured to integrate two execution standards and two safety standards. Type D, irrespective of the work sequence, described each execution standard in a random order, followed by randomly arranging the safety standards in the same manner.

3.2.3. Conducting the Experiment

The experiment involved a total of 660 participants, divided into four groups of 165 individuals each, based on the type of training material. These participants were workers from a large construction site located in Goduk-myeon, Pyeongtaek-si, Gyeonggi-do. The characteristics of each group were as follows:
  • Group A: trained with materials that simultaneously explain execution standards and safety standards according to the construction sequence, measuring the count of recalls.
  • Group B: trained with materials that explain execution standards and safety standards separately according to the construction sequence, measuring the count of recalls.
  • Group C: trained with materials that simultaneously explain execution standards and safety standards in a random order regardless of the construction sequence, measuring the count of recalls.
  • Group D: trained with materials that explain execution standards and safety standards separately in a random order regardless of the construction sequence, measuring the count of recalls.
As shown in Table 1, an examination of the demographic characteristics of the four groups reveals that, on average, 87% of the participants across the groups are male. Regarding age distribution, individuals in their 50s constituted the largest group at an average of 29%, followed in descending order by those in their 40s, 30s, 20s, and over 60s, with group C not including any participants over 60. In terms of construction experience, those with less than a year of experience constituted the largest portion at 34%, while those with over 10 years of experience accounted for 26%. An analysis of the composition of experimental subjects for the steel construction task, based on the experimental data, revealed that, on average, 86% of participants across all four groups had no prior work experience, showing a similar distribution. Regarding accident experience, the distribution was as follows: 63% in group A, 55% in group B, 64% in group C, and 66% in group D had no accident experience.
Table 1. Demographic characteristics by group.
There is no difference between groups and demographic characteristics (p > 0.05). This indicates that the four groups share similar demographic distributions, making them suitable for comparative analysis of number of recalls (recall rate) through experimental methods.

3.2.4. Control of Experiment

  • Training Method
To control for variables that could result in the training content not being delivered uniformly across groups due to the instructor’s psychological state, teaching skills, and environmental factors, the training materials for all four types were recorded with the same electronic voice narration. Narrations were based on identical scripts for each standard, with sequences altered by type to ensure uniform delivery of the information.
  • Training Duration
Excluding preparation and warm-up activities, the actual TBM duration was set at approximately 5–10 min. To ensure that the training duration for each group of subjects would be around five minutes, the training materials were produced as videos.
  • Response to Recall Test
This experiment focuses more on the extent to which the trained content is recalled rather than correctly answering the questions posed in the response sheet. To avoid data corruption due to forced responses at one’s discretion when participants cannot remember clearly, they were thoroughly instructed to mark “do not remember” if unsure. This instruction was reiterated in the response sheet to ensure awareness and appropriate responses.

3.2.5. Normality Test and Statistical Analysis

Normality was assessed through measures of kurtosis and skewness. To confirm the homogeneity of demographic characteristics across the four groups, a Chi-square test based on cross-tabulation was conducted. This test was used to determine whether the distribution of demographic variables was similar across groups, ensuring that any observed differences in recall rates could be attributed to training methods rather than individual background characteristics [80].
Statistical significance between the groups was determined using the p-value, with a conventional significance threshold set at 0.05. Independent sample t-tests were conducted to compare recall performance between two groups, while ANOVA was employed to identify differences in recall characteristics among three or more groups. Additionally, t-tests were used to examine differences in recall performance between specific training conditions (e.g., sequence-based vs. random), and ANOVA was used when comparing more than two conditions. A p-value less than 0.05 was considered statistically significant, following standard practices in behavioral science research [80].

4. Results

4.1. Comparison of Between Number of Recalls and Recall Rate in Each Group

The results of this experiment regarding the composition of TBM training materials and the number of recalls and average recall rates for each question are summarized in Table 2. Here, the number of recalls refers to the sum of number of recalls per question for 165 participants in each group, and the average recall rate represents the recall ratio per question.
Table 2. Number of recalls and recall rates by group.
In the delivery method based on the sequence of operations, when comparing the number of recalls for training materials that either integrated or separated execution standards and safety standards, separating the standards for training resulted in an average of 1.18 more recalls compared to integrated training. This indicates a capacity to recall approximately 15% more information.
Comparing the number of recalls for training materials that either integrated or separated execution standards and safety standards within a sequence-based delivery method revealed that training by separating the standards resulted in an average of 0.4 more recalls than integrated training, which corresponds to approximately 6% more information being recalled. Additionally, a rapid decrease in the 60% recall rate was observed for both group C and group D.
A reliability test was conducted to verify the likelihood of obtaining consistent measurements for the same questions if measured repeatedly, prior to analyzing these experimental results. The reliability of the responses to the experimental questions, as indicated by the Cronbach’s alpha coefficient, was 0.689, confirming reliability since the value is above 0.6.
Furthermore, descriptive statistical analysis was performed to examine the statistical characteristics of the variables. The average number of recalls was found to be 7.42 with a standard deviation of 3.280. Skewness and kurtosis values not exceeding an absolute value of 3 and 8, respectively, suggest that the variable can be considered to follow a normal distribution [81]. The review of normality for this research experiment’s data indicated that there were no issues with the univariate normality assumption regarding skewness and kurtosis.
As shown in Table 3, the average number of recalls of overall, group A, group B, group C, and group D were 7.42, 7.87, 9.05, 6.19, and 6.59, respectively. The relationship between groups and number of recalls was statistically significant (p < 0.001). Additionally, the average total recall rate was 49% for group A, 57% for group B, 39% for group C, 41% for group D, and 47% for overall. The relationship between groups and total recall rate was statistically significant (p < 0.001).
Table 3. Results of each group in recall experiment.

4.2. Comparison of Number of Recalls Based on Information Delivery Method

A comparison was made between the amount of information delivered through sequential order-based training materials and that of random order-based materials, examining the difference in number of recalls. T-test was employed for comparing the recall rates between sequential and random methods. As shown in Table 4, the average number of recalls for the sequence-based method was 8.46, while for the random method, it was 6.39, showing a statistically significant difference (p < 0.01).
Table 4. Number of recalls of two education methods by sequence and random.
Analyzing this in relation to cognitivist learning theories, it is inferred that the sequence of operations as a mediator in the information processing journey from sensory registration to short-term memory positively influenced attention and perception activities. To enhance workers’ cognition through pre-work safety training, organizing, providing safety training information according to the sequence of operations is advantageous for improving effectiveness.
This study also compared the number of recalls between integrated and separated configurations of execution standards and safety standards. Table 5 presents the average number of recalls for group A and group C, which were tested with integrated standards, was 7.03, while for group B and group D, tested with separated standards, the average number of recalls was 7.82, showing a statistically significant difference (p < 0.01). The recall rate for the groups tested with integrated standards (group A, group C) was 43.9%, and for those tested with separated standards (group B, group D), the recall rate was 48.9%, confirming that training through separated standards enhances cognitive effectiveness.
Table 5. Number of recalls of two education methods by integration and separation for two standards.

4.3. Comparison of Number of Recalls by Demographic Characteristics

No significant difference was found in number of recalls between gender and all groups (p > 0.05) (Table 6). As shown in group A, the average for men was 8.04 and 7.07 for women. In group B, the average recall for men was 9.14 and for women, it was 7.92. For group C, men had an average recall of 6.25, while women had an average of 5.82. In group D, the average recall for men was 6.62, and for women, it was 6.00.
Table 6. Number of recalls of each group by gender.
Table 7 shows total recalls by age. In group A, individuals in their 50s recalled more than those over 60, but no significant differences in number of recalls due to age were observed in the other groups. The average for individuals in their 20s was 7.03, in their 30s was 7.56, in their 40s was 8.08, in their 50s was 8.88, over 60s was 5.38, with an overall average of 7.87, showing a statistically significant difference in group A (p < 0.01). For group B, the average for those in their 20s was 8.87, in their 30s was 8.90, in their 40s was 8.71, in their 50s was 9.62, over 60s was 6.00, with an overall average of 9.05. In group C, the average for those in their 20s was 6.21, in their 30s was 5.71, in their 40s was 6.18, in their 50s was 6.58, with an overall average of 6.19. For group D, the average for those in their 20s was 6.32, in their 30s was 6.39, in their 40s was 6.67, in their 50s was 6.93, over 60s was 6.50, with an overall average of 6.59.
Table 7. Number of recalls of each group by age.
From Table 8, no significant differences in number of recalls based on experience were observed among all groups (p > 0.05).
Table 8. Number of recalls of each group by career in construction industry.
Table 9 shows total recalls by work experience. In group A, workers with experience using the steel construction training materials utilized in the experiment had a higher recall rate than those without experience, but no significant differences were found in other groups. Upon examining group A, the average recall rate for those with experience was 9.54, compared to 7.53 for those without, showing a statistically significant difference (p < 0.01). In group B, the average recall rate for those with experience was 9.24, while it was 9.01 for those without experience. For group C, the average recall rates were 6.28 for those with experience and 6.18 for those without. In group D, the averages were 6.57 for those with experience and 6.59 for those without, indicating no significant difference in recall rates based on experience within these groups.
Table 9. Number of recalls of each group by work experience.
In all four groups, there was no significant difference in recall rates based on accident experience (p > 0.05) (Table 10).
Table 10. Number of recalls of each group by accident experience.

5. Discussion

The findings demonstrate that the sequence-based delivery method significantly enhanced participants’ recall of safety training content compared to the randomly structured method. Participants in the sequence-based group recalled an average of 2.07 more items and achieved a 32% higher recall rate. This suggests that aligning training content with the actual work sequence enhances memory retention and comprehension.
Further analysis examined the effects of integrating versus separating execution and safety standards within each delivery method. Within the sequence-based method, separating the standards (Group B) led to a higher average recall count (9.05) and recall rate (56.6%) than integrating the standards (Group A), which had a recall count of 7.87 and a rate of 49.2%. This indicates that distinguishing between execution and safety standards facilitates better cognitive processing and information retention when training follows a task sequence.
Conversely, in the random delivery method, there was no statistically significant difference between the integrated (Group C) and separated (Group D) groups, with recall rates of 38.7% and 41.2%, respectively. The t-test confirmed this lack of significance (t = −1.219, p = 0.224), suggesting that separating standards is only effective when the training content is presented in a meaningful sequence.
No significant differences in recall performance were found across gender or work experience in any group. However, within group A, participants in their 50s recalled more items (8.88) than those aged 60 and above (5.38), indicating potential age-related cognitive differences in sequence-based training. Similarly, workers with prior experience using the specific steel construction training materials recalled more items (9.54) than those without such experience (7.53), highlighting the potential benefit of material familiarity on training effectiveness.
This study found that structuring training materials according to the work process and distinguishing between types of information influence the effectiveness of safety education for workers, providing practical guidance for pre-work training in the construction industry.

6. Conclusions

This study investigated the effectiveness of safety training materials that educate construction workers on execution and safety standards. The results demonstrated that organizing training content in alignment with the sequence of operations increased training effectiveness by approximately 32% compared to randomly structured content. This suggests that sequencing positively influences attention and perception during the cognitive processing of information. Furthermore, training that separated execution standards from safety standards showed approximately 15% greater effectiveness than integrated training, potentially due to the grouping principles outlined in Gestalt learning theory.
No significant differences were found among the four experimental groups in terms of individual characteristics, indicating the general applicability of these training strategies across different worker profiles.
These findings contribute to the body of knowledge on construction safety training by providing empirical evidence supporting the structured delivery of TBM content. To enhance the effectiveness of daily TBM sessions on construction sites, it is recommended to structure training materials according to task sequence and to clearly separate execution and safety standards. These approaches improve cognitive processing and overall training impact.
However, this study was limited by its short-term design and a participant group consisting solely of newly hired workers. Given that TBMs are conducted daily in real-world settings, future research should investigate long-term cognitive changes resulting from repeated exposure to training. Additionally, considering the pivotal role of experienced workers in construction, further studies should explore the effectiveness of TBM strategies for this demographic.

Author Contributions

Formal analysis, G.Y.K.; Visualization, G.Y.K. and Y.B.K.; Writing—original draft, G.Y.K.; Writing—review and editing, Y.B.K.; Investigation, Y.B.K., H.B. and J.Y.P.; Validation, H.B.; Methodology, H.B.; Conceptualization, H.K.K.; Data curation, H.K.K.; Resources, H.K.K.; Supervision, J.Y.P. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The data presented in this study are available on request from the corresponding authors.

Conflicts of Interest

Author Hyeong Keun Kim was employed by the company Samsung C&T Corporation. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Sánchez, F.A.S.; Peláez, G.I.C.; Alís, J.C. Occupational safety and health in construction: A review of applications and trends. Ind. Health 2017, 55, 210–218. [Google Scholar] [CrossRef] [PubMed]
  2. Hananpi, N.M.; Kamal, M.M.M.; Ismail, M.I.; Abdullah, I.A.P. Identifying Root Causes and Mitigation Measures of Construction Fall Accidents. Gading Bus. Manag. J. 2013, 17, 65–79. [Google Scholar]
  3. Sousa, V.; Almeida, N.M.; Dias, L.A. Risk-based management of occupational safety and health in the construction industry–Part 2: Quantitative model. Saf. Sci. 2015, 74, 184–194. [Google Scholar] [CrossRef]
  4. Adami, P.; Rodrigues, P.B.; Wodds, P.J.; Becerik-Gerber, B.; Soibelman, L.; Copur-Gencturk, Y.; Lucas, G. Effectiveness of VR-based training on improving construction workers’ knowledge, skills, and safety behavior in robotic teleoperation. Adv. Eng. Inform. 2021, 30, 101431. [Google Scholar] [CrossRef]
  5. Kim, J.; Lee, H.; Park, M.; Kwon, N. A System Dynamics Approach for Modeling Cognitive Process of Construction Workers’ Unsafe Behaviors. Korean J. Constr. Eng. Manag. 2017, 18, 38–48. [Google Scholar] [CrossRef]
  6. Sousa, V.; Almeida, N.M.; Dias, L.A. Risk-based management of occupational safety and health in the construction industry–Part 1: Background knowledge. Saf. Sci. 2014, 66, 75–86. [Google Scholar] [CrossRef]
  7. U.S. Bureau of Labor Statistics. Available online: https://www.bls.gov/charts/census-of-fatal-occupational-injuries/number-and-rate-of-fatal-work-injuries-by-industry.htm (accessed on 14 February 2024).
  8. Brown, S.; Brooks, R.D.; Dong, X.S. CPWR—The Center for Construction Research and Training; National Institute for Occupational Safety and Health. Nonfatal Injuries in Construction, 2003–2019. CPWR Data Bulletin, December 2020. Available online: https://stacks.cdc.gov/view/cdc/115417/cdc_115417_DS1.pdf (accessed on 21 February 2024).
  9. Eurostat. Available online: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Accidents_at_work_-_statistics_by_economic_activity (accessed on 14 February 2024).
  10. Jeschke, K.C.; Kines, P.; Rasmussen, L.; Andersen, L.P.; Dyreborg, J.; Ajslev, J.; Kabel, A.; Jensen, E.; Andersen, L.L. Process evaluation of a Toolbox-training program for construction foremen in Denmark. Saf. Sci. 2017, 94, 152–160. [Google Scholar] [CrossRef]
  11. Ministry of Employment and Labor. Analysis of Industrial Accident in South Korea 2021; Ministry of Employment and Labor: Sejong, Republic of Korea, 2022; pp. 1–626.
  12. Heinrich, H.W. Industrial Accident Prevention: A Scientific Approach, 2nd ed.; Mcgraw-Hill Book Company, Inc.: New York, NY, USA; London, UK, 1941; pp. 1–448. [Google Scholar]
  13. Törner, M.; Pousette, A. Safety in construction–a comprehensive description of the characteristics of high safety standards in construction work, from the combined perspective of supervisors and experienced workers. J. Saf. Res. 2009, 40, 399–409. [Google Scholar] [CrossRef] [PubMed]
  14. Soh, J.; Jeong, J.; Jeong, J.; Lee, J. Quantitative Risk Evaluation by Building Type Based on Probability and Cost of Accidents. Buildings 2023, 13, 327. [Google Scholar] [CrossRef]
  15. Tixier, A.J.P.; Hallowell, M.R.; Albert, A.; van Boven, L.; Kleiner, B.M. Psychological antecedents of risk-taking behavior in construction. J. Constr. Eng. Manag. 2014, 140, 04014052. [Google Scholar] [CrossRef]
  16. Hallowell, M.R. A Formal Model for Construction Safety and Health Risk Management. Ph.D. Thesis, Oregon State University, Corvallis, OR, USA, 2008. [Google Scholar]
  17. Ashuro, Z.; Zele, Y.T.; Kabthymer, R.H.; Diriba, K.; Tesfaw, A.; Alamneh, A.A. Prevalence of Work-Related Injury and Its Determinants among Construction Workers in Ethiopia: A Systematic Review and Meta-Analysis. J. Environ. Public Health 2021, 2021, 9954084. [Google Scholar] [CrossRef] [PubMed]
  18. Kumar, D.; Bhattacharjee, R.M. Reducing workplace unsafe behaviour using risk classification, profiling, risk tolerance approach. Heliyon 2023, 9, e13969. [Google Scholar] [CrossRef]
  19. Health and Safety Executive HSE. Strategies to Promote Safe Behavior as Part of a Health and Safety Management System; Contract Research; Rep. No. 430; SE Books: Merseyside, UK, 2002. Available online: https://webarchive.nationalarchives.gov.uk/ukgwa/20230103132016 (accessed on 16 April 2025).
  20. Ni, G.; Lv, L.; Wang, S.; Miao, X.; Fang, Y.; Liu, Q. Formation Mechanism and Dynamic Evolution Laws About Unsafe Behavior of New Generation of Construction Workers Based on China’s Construction Industry: Application of Grounded Theory and System Dynamics. Front. Psychol. 2022, 13, 888060. [Google Scholar] [CrossRef]
  21. Li, P.; He, Y.; Li, Z. Study on Influencing Factors of Construction Workers’ Unsafe Behavior Based on Text Mining. Front. Psychol. 2022, 13, 886390. [Google Scholar] [CrossRef] [PubMed]
  22. Peng, L.; Chan, A.H.S. Exerting Explanatory Accounts of Safety Behavior of Older Construction Workers within the Theory of Planned Behavior. Int. J. Environ. Res. Public Health 2019, 16, 3342. [Google Scholar] [CrossRef] [PubMed]
  23. Reason, J. Human Error; Cambridge University Press: Cambridge, UK, 1990; pp. 1–393. [Google Scholar]
  24. Suraji, A.; Duff, A.R.; Peckitt, S.J. Development of causal model of construction accident causation. J. Constr. Eng. Manag. 2001, 127, 337–344. [Google Scholar] [CrossRef]
  25. Jiang, Z.; Fang, D.; Zhang, M. Understanding the Causation of Construction Workers’ Unsafe Behaviors Based on System Dynamics Modeling. J. Manag. Eng. 2015, 31, 1–14. [Google Scholar] [CrossRef]
  26. Sa, J.; Seo, D.-C.; Choi, S.D. Comparison of risk factors for falls from height between commercial and residential roofers. J. Saf. Res. 2009, 40, 1–6. [Google Scholar] [CrossRef] [PubMed]
  27. Sulzer-Azaroff, B. Behavioral approaches to occupational health and safety. In Handbook of Organizational Behavior Management; Frederiksen, L.W., Ed.; John Wiley & Sons, Inc.: New York, NY, USA, 1982; pp. 505–538. [Google Scholar]
  28. Moon, K.; Lee, K.; Lee, J.; Oah, S. The effect of behavior based safety (BBS) program on safety climate and safety behaviors: A field study. Korean J. Ind. Organ. Psychol. 2012, 25, 349–372. [Google Scholar] [CrossRef]
  29. Austine, J.; Kessler, M.L.; Riccobono, J.E.; Bailey, J.S. Using Feedback and Reinforcement to Improve the Performance and Safety of a Roofing Crew. J. Organ. Behav. Manag. 1996, 16, 49–65. [Google Scholar] [CrossRef]
  30. Fox, D.K.; Hopkins, B.L.; Anger, W.K. The long-term effects of a token economy on safety performance in open-pit mining. J. Appl. Behav. Anal. 1987, 20, 215–224. [Google Scholar] [CrossRef]
  31. Reber, A.R.; Wallin, J.A.; Chhokar, J.S. Improving Safety Performance with Goal Setting and Feedback. Hum. Perform. 1990, 3, 51–61. [Google Scholar] [CrossRef]
  32. Choudhry, R.M.; Fang, D.; Mohamed, S. Developing a Model of Construction Safety Culture. J. Manag. Eng. 2007, 23, 207–212. [Google Scholar] [CrossRef]
  33. Xia, N.; Xie, Q.; Griffin, M.A.; Ye, G.; Yuan, J. Antecedents of safety behavior in construction: A literature review and an integrated conceptual framework. Accid. Anal. Prev. 2020, 148, 105834. [Google Scholar] [CrossRef] [PubMed]
  34. Fang, D.; Zhao, C.; Zhang, M. A Cognitive Model of Construction Workers’ Unsafe Behaviors. J. Constr. Eng. Manag. 2016, 142, 9. [Google Scholar] [CrossRef]
  35. Dejoy, D.M. Behavior change versus culture change: Divergent approaches to managing workplace safety. Saf. Sci. 2005, 43, 105–129. [Google Scholar] [CrossRef]
  36. Gould, F.; Joyce, N. Construction Project Management, 3rd ed.; Pearson Prentice Hall: Upper Saddle River, NJ, USA, 2009; pp. 1–384. [Google Scholar]
  37. Shin, M.; Lee, H.-S.; Park, M.; Moon, M.; Han, S. A system dynamics approach for modeling construction workers’ safety attitudes and behaviors. Accid. Anal. Prev. 2014, 68, 95–105. [Google Scholar] [CrossRef] [PubMed]
  38. Jung, J.-Y.; Yu, H.-J. A Study for Construction Workers’ Safety Consciousness and Safety Education Activation Methods. J. Korean Soc. Saf. 2013, 28, 47–51. [Google Scholar] [CrossRef]
  39. Robson, L.S.; Stephenson, C.M.; Schulte, P.A.; Amick, B.C.; Irvin, E.L.; Eggerth, D.E.; Chan, S.; Bielecky, A.R.; Wang, A.M.; Heidotting, T.L.; et al. A systematic review of the effectiveness of occupational health and safety training. Scand. J. Work. Environ. Health 2012, 38, 193–208. [Google Scholar] [CrossRef]
  40. Haupt, T.C. A Study of management attitudes to a performance approach to construction worker safety. J. Constr. Res. 2003, 4, 87–100. [Google Scholar] [CrossRef]
  41. Becker, P.; Morawetz, J. Impacts of Health and Safety Education: Comparison of Worker Activities Before and After Training. Am. J. Ind. Med. 2004, 46, 63–70. [Google Scholar] [CrossRef] [PubMed]
  42. Rice, S.P.M.; Rimby, J.; Hurtado, D.A.; Gilbert-Jones, I.; Olson, R. Does sending Safety Toolbox Talks by text message to Residential Construction Supervisors increase Safety Meeting Compliance? Occup. Health Sci. 2022, 6, 313–332. [Google Scholar] [CrossRef]
  43. Zohar, D. Safety climate in industrial organizations: Theoretical and applied implications. J. Appl. Psychol. 1980, 65, 96–102. [Google Scholar] [CrossRef] [PubMed]
  44. Hoffmann, D.A.; Morgeson, F.P. Safety-related behavior as a social exchange: The role of perceived organizational support and leader-member exchange. J. Appl. Psychol. 1999, 84, 286–296. [Google Scholar] [CrossRef]
  45. Kim, M.-J. The Improvement Plan of Construction Safety Education for Accident Prevention in Construction Sites. Master’s Thesis, Dongguk University, Seoul, Republic of Korea, 2003. [Google Scholar]
  46. Korea Statistical Information Service. Available online: https://kosis.kr/statHtml/statHtml.do?orgId=380&tblId=DT_380004_C18_1&vw_cd=MT_ZTITLE&list_id=C_13_001_001&seqNo=&lang_mode=ko&language=kor&obj_var_id=&itm_id=&conn_path=MT_ZTITLE (accessed on 19 September 2023).
  47. Varley, F.D.; Noldt, C.M.K. Developing Toolbox Training Materials for Mining. In Strategies for Improving Miners’ Training; Peters, R.H., Ed.; DHHS (NIOSH) Publication No. 2002–156; U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health: Pittsburgh, PA, USA, 2002; pp. 27–34. [Google Scholar]
  48. Kaskutas, V.; Jaegers, L.; Dale, A.M.; Evanoff, B.A. Toolbox Talks: Insights for Improvement. Prof. Saf. 2016, 61, 33–37. [Google Scholar]
  49. Oh, M.-H.; Park, T.-H.; Park, Y.-J.; Son, K.-Y.; Ahn, S.-J.; Kim, T.-H. An Assessment of Safety Education Contents Propriety Analyzing Accident Types by Work Classification in Construction Sites. J. Archit. Inst. Korea Struct. Constr. 2014, 30, 131–139. [Google Scholar] [CrossRef]
  50. Lim, D.; Lee, D.; Moon, Y.; Woo, S.; Jeong, A.; Park, J. Exploring Safety Management Approaches for Construction Sites Using Smart Devices. In Proceedings of the 2024 10th International Conference on Condition Monitoring and Diagnosis (CMD), Gangneung, Republic of Korea, 20–24 October 2024; pp. 485–488. [Google Scholar]
  51. Kwon, S.J.; Choi, S.W.; Lee, E.B. Hazard Identification and Risk Assessment During Simultaneous Operations in Industrial Plant Maintenance Based on Job Safety Analysis. Sustainability 2024, 16, 9277. [Google Scholar] [CrossRef]
  52. Construction Technology Promotion Act, Enforcement Decree, Article 103, Safety Education, Republic of Korea, 2022. Available online: https://elaw.klri.re.kr/kor_service/lawView.do?hseq=68275&lang=KOR (accessed on 16 April 2025).
  53. Gagné, R.M.; Briggs, L.J.; Wager, W.W. Principles of Instructional Design, 4th ed.; Harcourt Brace College Publisher: Belmont, CA, USA, 1979; pp. 1–365. [Google Scholar]
  54. Bandura, A. Self-efficacy: Toward a unifying theory of behavior change. Psychol. Rev. 1977, 84, 191–215. [Google Scholar] [CrossRef] [PubMed]
  55. Cho, J.Y.; Kang, M.S. Significance and Directions of the Instruction which Applied the Advanced Organizer: Based on the Brain Science Research. J. Elem. Educ. 2014, 27, 1–25. [Google Scholar]
  56. Synn, M.H. Education Psychology; Hakjisa: Seoul, Republic of Korea, 2010; pp. 1–432. [Google Scholar]
  57. Kang, D.I.; Kim, C.-S.; Choi, H.C. Audience Schema and the Effects of Mass Communications. Sungkok Nonchong 1991, 22, 103–157. [Google Scholar]
  58. Foley, H.J.; Matlin, M.W. Sensation and Perception, 5th ed.; Psychology Press: New York, NY, USA, 2010; pp. 1–516. [Google Scholar]
  59. Rock, I.; Palmer, S. The Legacy of Gestalt Psychology. Sci. Am. 1990, 263, 84–91. [Google Scholar] [CrossRef] [PubMed]
  60. Shuell, T.J. Cognitive Conceptions of Learning. Rev. Educ. Res. 1986, 56, 411–436. [Google Scholar] [CrossRef]
  61. Reed, S.K. Cognition: Theory and Applications, 7th ed.; Thomson Wadsworth: Belmont, CA, USA, 2007; pp. 1–426. [Google Scholar]
  62. Eggen, P.; Kauchak, D. Strategies for Teachers: Teaching Content and Thinking Skills, 3rd ed.; Ally and Bacon: Boston, MA, USA, 2001; pp. 1–368. [Google Scholar]
  63. Atkinson, R.C.; Shiffrin, R.M. Human memory: A proposed system and its control processes. Psychol. Learn. Motiv. 1968, 2, 89–195. [Google Scholar]
  64. Akturk, A.O.; Sahin, I. Literature Review on Metacognition and its Measurement. Procedia Soc. Behav. Sci. 2011, 15, 3731–3736. [Google Scholar] [CrossRef]
  65. Bonner, J. Implications of cognitive theory for instructional design: Revisited. Educ. Commun. Technol. J. 1988, 36, 3–14. [Google Scholar] [CrossRef]
  66. Sweller, J.; Merrienboer, J.J.G.; Paas, F.G.W.C. Cognitive Architecture and Instructional Design. Educ. Psychol. Rev. 1998, 10, 251–296. [Google Scholar] [CrossRef]
  67. Sweller, J. Cognitive load theory, learning difficulty, and instructional design. Learn. Instr. 1994, 4, 295–312. [Google Scholar] [CrossRef]
  68. Sweller, J. Instructional Design in Technical Areas; Australian Council for Educational: Melbourne, Australia, 1999; pp. 1–192. [Google Scholar]
  69. Jalani, N.H.; Sern, L.C. The Example-Problem-Based Learning Model: Applying Cognitive Load Theory. Procedia Soc. Behav. Sci. 2015, 195, 872–880. [Google Scholar] [CrossRef]
  70. Garrett, J.W.; Teizer, J. Human Factors Analysis Classification System Relating to Human Error Awareness Taxonomy in Construction Safety. J. Constr. Eng. Manag. 2009, 135, 754–763. [Google Scholar] [CrossRef]
  71. Ebbinghaus, H. Memory: A Contribution to Experimental Psychology; Columbia University: Dover, Japan; New York, NY, USA, 1962; pp. 1–123. [Google Scholar]
  72. Sun, Z.; Zhang, C.; Tang, P. Simulation-Based Optimization of Communication Protocols for Reducing Delays during Nuclear Power Plant Outages. In Proceedings of the Construction Research Congress 2018: Infrastructure and Facility Management, CRC 2018, New Orleans, LA, USA, 2–4 April 2018. [Google Scholar]
  73. Katz, A.; Shtub, A.; Roguin, A. Minimizing Ionizing Radiation Exposure in Invasive Cardiology Safety Training for Medical Doctors. J. Nucl. Eng. Radiat. Sci. 2017, 3, 030905. [Google Scholar] [CrossRef]
  74. Kovacs, G.; Bullock, G.; Ackroyd-Stolarz, S.; Cain, E.; Petrie, D. A randomized controlled trial on the effect of educational interventions in promoting airway management skill maintenance. Ann. Emerg. Med. 2000, 36, 301–309. [Google Scholar] [CrossRef] [PubMed]
  75. Raffel, G. The effect of recall on forgetting. J. Exp. Psychol. 1934, 17, 828–838. [Google Scholar] [CrossRef]
  76. Ministry of Land, Infrastructure and Transport. Construction Technology Promotion Act; Republic of Korea, 2022. Available online: https://elaw.klri.re.kr/eng_service/lawView.do?hseq=60987&type=eng (accessed on 16 April 2025).
  77. Ministry of Employment and Labor. Occupational Safety and Health Act; Republic of Korea, 2023. Available online: https://elaw.klri.re.kr/eng_service/lawView.do?hseq=66210&type=eng (accessed on 16 April 2025).
  78. Mathy, F.; Feldman, J. What’s magic about magic numbers? Chunking and data compression in short-term memory. Cognition 2012, 122, 346–362. [Google Scholar] [CrossRef] [PubMed]
  79. Miller, G.A. The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychol. Rev. 1956, 63, 81–97. [Google Scholar] [CrossRef]
  80. McHugh, M.L. The Chi-square test of independence. Biochem. Medica 2013, 23, 143–149. [Google Scholar] [CrossRef]
  81. Kline, R.B. Principles and Practice of Structural Equation Modeling, 2nd ed.; The Guilford Press: New York, NY, USA, 2005; pp. 1–366. [Google Scholar]
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.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Multi-Viewpoint Assessment of Urban Waterfront Skylines: Fractal and Spatial Hierarchy Analysis in Shanghai
Previous
Evaluation and Optimization of Urban Street Spatial Quality Based on Street View Images and Machine Learning: A Case Study of the Jinan Old City
Next