Factors Affecting the Well-Being at Work and Risk Perception of Construction Workers: A Validated Interpretative Structural Modeling (VISM) Approach

No.

Reviewers’ comments

Responses

Reviewer 1

1.        

Does the introduction provide sufficient background and include all relevant references? – Can be improved

Thanks for your supportive comment.

We added some relevant references in order to address this item. We hope it is suitable.

2.        

Are all the cited references relevant to the research? – Yes

Thanks for your supportive comment.

3.        

Is the research design appropriate? – Must be improved

Thanks for your supportive comment.

In order to address your questions, we have made changes to the introduction to highlight the research question, as can be seen in the file.

We hope it is suitable.

4.        

Are the methods adequately described? – Must be improved

Thanks for your supportive comment.

In order to address your and the other reviewers' questions, we have made significant revisions to the methods section, as can be seen in the file.

We hope it is suitable.

5.        

Are the results clearly presented? – Can be improved

Thanks for your supportive comment.

In order to address your and the other reviewers' questions, we have made significant revisions to the results section, as can be seen in the file.

We hope it is suitable.

6.        

Are the conclusions supported by the results? – Can be improved

Thanks for your comment.

We made major revisions in the conclusion section. We hope it is suitable.

7.        

Why use the Validated Interpretative Structural Modeling (VISM) approach? How is this analysis model different from others? In other words, can the research purpose of this article be accomplished without using this analysis model ?

Thanks for your comment.

To address your query, it was added the following discussion in the topic “3. Methoodology”: “The VISM was chosen because it allows the construction of conceptual models with strong statistical support from a reduced number of interviews, even in situations where the sample exhibits non-normal data distribution [71], as is the case in this re-search.

Furthermore, the relationship between well-being and the perception of occupational accident risks is still not well understood due to the complexity of these interactions. In this context, the VISM emerges as a technique capable of combining the explanatory power for complex phenomena of Interpretative Structural Modeling (ISM) with the statistical validity of Structural Equation Modeling – Partial Least Squares (PLS-SEM), enabling the creation of complex models with strong statistical support, even in non-linear models containing circular and/or cross relations among constructs [71]”.

We hope it is suitable.

8.        

What are the important characteristics or differences of the working environment and conditions in Brazil? This is an empirical investigation of background environmental factors that helps to understand the local working environment conditions.

Thanks for your comment.

The entire topic 2.2 was thoroughly revised with the goal of addressing your inquiry more clearly.

We hope it is suitable.

9.        

Another topic of this article is risk perception. In the literature review section, another word "safety climate" is used. Are the two concepts the same? Or is there a difference? How is it related to well-being at work?

Thanks for your comment.

The following paragraphs was added in section 2.3 to address your query: “Within this perception of safety in the work environment lies the worker's perception of accident risks, serving as one of the factors influencing the overall safety climate perception within the organization [65]” and “In this context, conceptually, this research explores factors related to workplace well-being and safety climate, with the perception of workplace accident risks being one of the components encompassed within the safety climate concept”.

We hope it is suitable.

10.     

Why were the cities of Belém, Ananindeua, Castanhal and Abaetetuba chosen as sample locations? What are the questionnaire implementation time, distribution and number of valid samples? Are there differences in the status and conditions of the working environment in these cities?

Thanks for your comment.

It was added the following text in the last paragraph of chapter 3.1: “These cities also had the highest number of ongoing construction projects in the state during the study period, between April and May 2019. Moreover, these cities shared the same labor relations, being governed by the same employers' and workers' unions and adhering to the same employment contract statutes. To ensure data quality, participants were randomly selected based on their availability and willingness to participate”.

Aditionaly, was obtained 376 valid responses. This information are presented in topic 4.1.

We hope it is suitable.

11.     

What is the content of the questionnaire? The recommendations are presented in the appendix and how reliable and valid is it?

Thanks for your comment.

Certainly, the content of the questionnaire is contained in Table 1. As it solely displays indicators derived from questionnaires developed by recognized researchers in the field, we deemed it unnecessary to include it again in the appendices.

Thank you for your understanding.

12.     

Frankly speaking, the conclusion of this article is not much different from the general knowledge, and there have been many related studies to prove it. So are there some unique findings worth explaining? Let them become the most important discovery and contribution of this article.

Thanks for your comment.

The entire conclusion was thoroughly revised with the goal of better explain the paper contributions, implications and limitations.

We hope it is suitable.

No.

Reviewers’ comments

Responses

Reviewer 2

1.        

Does the introduction provide sufficient background and include all relevant references? – Can be improved

Thanks for your supportive comment.

We added some relevant references in order to address this item. We hope it is suitable.

2.        

Are all the cited references relevant to the research? – Can be improved

Thanks for your supportive comment.

We added some relevant references in order to address this item. We hope it is suitable.

3.        

Is the research design appropriate? – Can be improved

Thanks for your supportive comment.

In order to address your questions, we have made changes to the introduction to highlight the research question, as can be seen in the file.

We hope it is suitable.

4.        

Are the methods adequately described? – Can be improved

Thanks for your supportive comment.

In order to address your and the other reviewers' questions, we have made significant revisions to the methods section, as can be seen in the file.

We hope it is suitable.

5.        

Are the results clearly presented? – Can be improved

Thanks for your supportive comment.

In order to address your and the other reviewers' questions, we have made significant revisions to the results section, as can be seen in the file.

We hope it is suitable.

6.        

Are the conclusions supported by the results? – Can be improved

Thanks for your comment.

We made major revisions in the conclusion section. We hope it is suitable.

7.        

You need to make some statements related to your research by insert in the first part the research question(s). Then, in the Introduction you need to rethink and rewrite the last paragraph (lines 98-106). It is difficult to understand what was meant here. Basically, you have through the entire manuscript some other similar situations and therefore you need to carry out a full check in terms of spelling, grammar and sentence construction.

Thanks for your comment.

The paragraph from lines 98 to 106 has been removed. Additionally, the research question was emphasized in the last paragraph of the "1. Introduction" chapter. “In this sense, from the perspective of construction workers in the Brazilian Amazon, what are the workplace well-being factors that affect their perception of accident risks?”

We hope it is suitable.

8.        

A useful info could be included in the subsection 2.3 where you could enter a table with some incidents happened in the civil construction sector in the last years in Brazil, for instance (entering some statistical info). 

Thanks for your comment.

The table 1 was added aiming to address some number about accidents in the Brazilian construction industry.

We hope it is suitable.

9.        

Then, related to the methodology I would like to ask you to develop the choice made for the technique Validated Interpretive Structural Modeling. Could you present also a comparison of similar method and your decision to integrate two modelling methods: Interpretative Structural Modeling and Structural Equation Modeling – Partial Least Square - Structural Equation Modeling. I ask for this in order to emphasize the pros and the cons of the approach taken. 

Thanks for your comment.

The following discussion was added to the "3. Methodology" section: “The VISM was chosen because it allows the construction of conceptual models with strong statistical support from a reduced number of interviews, even in situations where the sample exhibits non-normal data distribution [71], as is the case in this re-search.

Furthermore, the relationship between well-being and the perception of occupational accident risks is still not well understood due to the complexity of these interactions. In this context, the VISM emerges as a technique capable of combining the explanatory power for complex phenomena of Interpretative Structural Modeling (ISM) with the statistical validity of Structural Equation Modeling – Partial Least Squares (PLS-SEM), enabling the creation of complex models with strong statistical support, even in non-linear models containing circular and/or cross relations among constructs [71]”.

We hope it is suitable.

10.     

At the same time, it is necessary to detail the investigations carried out. Who designed the questionnaires (1 and 2)? What is the difference in approach and content between the first and the second investigation? When did the two investigations take place (in what time periods, at what time interval between them?)? What was the time interval dedicated to the two investigations (number of days or weeks, etc.?)? How were the subjects selected and what was the response rate obtained?

Thanks for your comment.

To emphasize how the first survey was conducted, the following paragraph was added in Section 3.3: “Once the indicators were identified through a systematic literature review, they were measured in the first survey of the research. This survey was conducted using a questionnaire developed based on models adopted by various renowned authors in the research field, as indicated in Table 1, the goal was to maintain them entirely in their original form to prevent distortions in their application and the responses obtained.” And the following passage was added to the last paragraph: “These cities also had the highest number of ongoing construction projects in the state during the study period, between April and May 2019. Moreover, these cities shared the same labor relations, being governed by the same employers' and workers' unions and adhering to the same employment contract statutes. To ensure data quality, participants were randomly selected based on their availability and willingness to participate”.

To emphasize how the second survey was conducted, the following paragraph was added in Section 3.3: “To develop the SSIM, a second questionnaire was created and emailed to accredited experts in the field of occupational safety and construction engineering in Brazil, including researchers and on-site engineers. The snowball sampling technique was employed for data collection, wherein the questionnaire was sent to one person, who then recommended other individuals to participate based on the sample criteria, until the research requirements were met”

We hope it is suitable.

11.     

Is it relevant to say that 5 people (even with great experience in the field) can generate, in the case of survey 2, a useful and adequate feedback for the research objectives?

Thanks for your comment.

As discussed in the text, the number of responses was comparable to that of other studies published in reputable journals. Furthermore, to address your query, the following highlighted passage was added to the first paragraph of Section 4.3: “As a result, five responses were obtained, a number equivalent to other surveys [71,78,79], and deemed sufficient, given that potential relationships would still be tested for statistical significance in the study”.

We hope it is suitable.

12.     

On the other hand, was there a survey data cleaning process for processing the data from the questionnaires applied in the two surveys? What were the specific steps followed for clean data? 

Thanks for your comment.

The complete dataset was utilized, as the questionnaires were administered face-to-face, ensuring that all participants responded to the questionnaire in its entirety.

We hope it is suitable.

13.     

Also, in order to understand the financial values, it is necessary to present an exchange rate R$ - USD or Euro.

Thanks for your comment.

Os valores em R$ foram convertidos para US$.

14.     

In the final part, additions are needed related to the theoretical and practical implications of the research, the main limitations of the proposed approach etc.

Thanks for your comment.

The entire conclusion was thoroughly revised with the goal of better explain the paper implications, limitations and contributions.

We hope it is suitable

15.     

The conclusions are formulated in a summary manner without presenting some important aspects generated by the analysis carried out.

No.

Reviewers’ comments

Responses

Reviewer 3

1.        

Does the introduction provide sufficient background and include all relevant references? – Yes

Thanks for your supportive comment.

2.        

Are all the cited references relevant to the research? – Yes

Thanks for your supportive comment.

3.        

Is the research design appropriate? – Can be improved

Thanks for your supportive comment.

In order to address your questions, we have made changes to the introduction to highlight the research question, as can be seen in the file.

We hope it is suitable.

4.        

Are the methods adequately described? – Can be improved

Thanks for your supportive comment.

In order to address your and the other reviewers' questions, we have made significant revisions to the methods section, as can be seen in the file.

We hope it is suitable.

5.        

Are the results clearly presented? – Yes

Thanks for your supportive comment.

6.        

Are the conclusions supported by the results? – Yes

Thanks for your supportive comment.

7.        

Line98-106 is not related to the introduction

Thanks for your comment.

The paragraph from lines 98 to 106 has been removed.

8.        

In Line138 the author states that the construction industry is unhealthy and cites the corresponding reasons, suggesting that more reasons be shown to support the points.

Thanks for your comment.

In response to your inquiry, we have added the following text to the passage: “In general, the construction industry is notably hazardous due to high numbers of on-site or commuting accidents [51]. Additionally, work-related illnesses in construction rank among the highest across all occupational groups [26].”

We hope it is suitable.

9.        

Iine265, the author mentions a pre-defined answer, is this a standard answer? Please provide further explanation or pictures.

Thanks for your comment.

To answer your query and emphasize how the first survey was conducted, the following paragraph was added in Section 3.3: “Once the indicators were identified through a systematic literature review, they were measured in the first survey of the research. This survey was conducted using a questionnaire developed based on models adopted by various renowned authors in the research field, as indicated in Table 1, the goal was to maintain them entirely in their original form to prevent distortions in their application and the responses obtained.”

We hope it is suitable.

10.     

What does the constructs refer to in 3.2, and provide a summary statement accordingly

Thanks for your comment.

In response to your requests, the explanation regarding the creation of constructs has been enhanced. The following text has been added: “First, the indicators were grouped according to the literature review conducted in Section 3.1, forming the constructs. Subsequently, confirmatory factor analysis was conducted to assess their coherence.”

We hope it is suitable.

11.     

In 3.5, consider adding images or tables for clearer presentation.

Thanks for your comment.

Unfortunately, we were unable to depict the procedure in a figure in this section. Hence, we have opted to explain it textually, as can be observed from the added text below: “In this method, first, an SSIM is constructed containing the valid relationships between the constructs obtained in Section 3.4. Then, the influences between constructs are translated into relationships and represented by arrows. Finally, the constructs and their relationships are organized into a structure containing the model's constructs and their interconnections.”

We hope it is suitable.

12.     

Line332-339, do different groups of respondents affect the results?

Thanks for your comment.

In order to address your question, the following text was added: “Despite the group being heterogeneous, no significant variations in the obtained responses were observed.”

We hope it is suitable.

13.     

And is it random to select groups?

Thanks for your comment.

In order to address your question, the following text was added to the ultimate paragraph of section 3.1: “To ensure data quality, participants were randomly selected based on their availability and willingness to participate”.

We hope it is suitable.

14.     

In 4.2, there is a lack of explanation of the indicators in Table 2, such as SW WBC.

Thanks for your comment.

To address your query, the next note was added in table 3: “** The abbreviations in the columns represent, in sequence, the constructs listed in the rows. For example, SW stands for Stress at Work, and ARP stands for Accident Risk Perception.”

We hope it is suitable.

15.     

How is the cross-loadings in Table 3 obtained? Please give further explanation

Thanks for your comment.

To address your query, the next text was added: “In Table 4, we can observe the cross-loadings between the constructs obtained through the measurement model of PLS-SEM.”

We hope it is suitable.

16.     

What is the specific method of statistical testing in Line377-378?

Thanks for your comment.

To address your query, the next text was added: “by bootstrapping procedure of the PLS-SEM”.

We hope it is suitable.

17.     

Please explain Figure 2

Thanks for your comment.

As per your request in query 11, the procedure for creating this figure has been detailed in the methodology section. Its content, on the other hand, is elaborated in the subsequent discussion section. To highlight this, we have added a sentence emphasizing this fact after the figure 2 title: “After creating the model using the VISM methodology, its findings and contributions are elaborated upon in the following discussion section”.

We hope it is suitable.

No.

Reviewers’ comments

Responses

Reviewer 1

1.        

Does the introduction provide sufficient background and include all relevant references? – Yes

Thanks for your supportive comment.

2.        

Are all the cited references relevant to the research? – Yes

Thanks for your supportive comment.

3.        

Is the research design appropriate? – Can be improved

Thanks for your supportive comment.

We made minors changes all long the document aiming to clarify the research aims, like substituting the word “safety” by “perception of risk of accidents” (last paragraph of section “1. Introduction”), referring to the terminology adopted in other sections.

We hope it is suitable.

4.        

Are the methods adequately described? – Yes

Thanks for your supportive comment.

5.        

Are the results clearly presented? – Yes

Thanks for your supportive comment.

6.        

Are the conclusions supported by the results? – Yes

Thanks for your comment.

7.        

The authors of this article have provided clarifications and corrections to the comments and suggestions in a short period of time, which is very admirable. 

We are deeply grateful for your comments and contributions. We are fully committed to addressing your and other revisers requirements. In this regard, we hope the current adjustments align with the demands.

8.        

Please refer to the journal requirements to adjust some of the formatting in this article.

Thanks for your comment.

We have carefully reviewed the journal requirements and made necessary adjustments to the formatting in the article. Especially witch related to author declaration, funding, figures and tables titles.

We hope it is suitable.

9.        

Recheck the abstract for consistency with the revised conclusions.

Thanks for your comment.

We rewrite the abstract. As can be seen above: “Despite increased safety investments in the civil construction sector, high accident rates persist, often due to workers' intentional unsafe behaviours influenced by poorly understood psychological factors. This study utilized Validated Interpretative Structural Modeling (VISM), an innovative technique, to investigate factors impacting the well-being and risk perception of construction workers in the Brazilian Amazon. The VISM model was developed through five steps: (1) identification of indicators; (2) definition of constructs; (3) identification of relationships between constructs; (4) validation of these relationships; and (5) creation of the model. The results underscore the effectiveness of VISM in crafting complex models with robust statistical validity, even in researches with sample limitations and intricate or circular relationships among constructs. The research identified various factors in this phenomenon, such as the importance given by management to workplace safety, job dissatisfaction, work-related stress, intention to turnover, work-family balance, alcohol and cigarette use, physical health, perception of accident risks at work, and education and training in workplace safety. Other significant contribution lies in the model's ability to discern both direct and indirect relationships among these well-being factors. This insight enables the formulation of strategies to enhance worker well-being and reduce workplace accidents, particularly those stemming from intentional unsafe behaviours.”

We hope it is suitable.

No.

Reviewers’ comments

Responses

Reviewer 2

1.        

Does the introduction provide sufficient background and include all relevant references? – Can be improved

Thanks for your comment.

We have made the adjustments that we deemed possible to address the improvement suggestions.

We hope it is suitable.

2.        

Are all the cited references relevant to the research? – Can be improved

Thanks for your comment.

We have made the adjustments that we deemed possible to address the improvement suggestions.

We hope it is suitable.

3.        

Is the research design appropriate? – Can be improved

Thanks for your supportive comment.

We made minors changes all long the document aiming to clarify the research aims, like substituting the word “safety” by “perception of risk of accidents” (last paragraph of section “1. Introduction”), referring to the terminology adopted in other sections.

We hope it is suitable.

4.        

Are the methods adequately described? – Can be improved

We have made the adjustments that we deemed possible to address the improvement suggestions, and we appreciate your contribution through the comments.

We hope it is suitable.

5.        

Are the results clearly presented? – Can be improved

Thanks for your comment.

We have made the adjustments that we deemed possible to address the improvement suggestions.

We hope it is suitable.

6.        

Are the conclusions supported by the results? – Can be improved

Thanks for your comment.

7.        

Your comments, additions and remarks are satisfactory. To be honest, I expected the answers related to the methodology to be more detailed and with a focus on the requirements developed in the review. Here there was still room for additional clarifications and clarifying explanations.

Thanks for your comment.

We carefully considered your suggestions and revised all the methodology section, adding text in order to better explain the work, like you can see:

(section 3.1)” As a result, 26 indicators were identified, as shown in Table 3. The next step involved measuring these indicators.[…] To ensure response variability consistent with PLS-SEM, a 5-point Likert scale was adopted, where 1 = strongly disagree, 2 = disagree, 3 = partially agree, 4 = agree, and 5 = strongly agree.”;

(section 3.2) table 2 was added explaining the Confirmatory factor analysis rules;

(section 3.4) “To achieve this, linear models will be constructed containing the potential relationships between constructs and then tested for statistical significance using bootstrapping procedures, following Santana et al.'s [78] guidelines. According to the author, we should adopt the following five conditions for the creation and testing of linear models used to validate relationships between constructs:

1.         The linear variations developed should show as many relations between constructs as possible.

2.         Possible unsupported relations should be deleted;

3.         A good test of a possible relation is one, in which, in the same linear variation all existing relations between the constructs that precede it are tested;

4.         If possible, as many linear variations should be developed as the possible relations to be tested allow, otherwise, and there is no clear limit on the maximum number of linear variations to be tested, a sufficient number of linear variations should be tested (as rule 5 shown below);

5.         A sufficient number of linear variations developed is one in which each relation is tested a number of times equal to the number of constructs. For example, if a model has ten constructs, each possible relation should be tested at least ten times.”

We hope it is suitable.

8.        

A useful info could Comments on the Quality of English Language: Final checks: spelling, grammar and sentence construction.

Thanks for your comment.

We carefully considered your suggestions and revised the entire document regarding spelling, grammar, and sentence construction.

We hope it is suitable.