Meta-Analysis of Studies on Accident Contributing Factors in the Greek Construction Industry
Round 1
Reviewer 1 Report
1. In the first section “which contains the results of the systematic literature review and content analysis necessary”. However, there is no result of content analysis in Section 3.
2. In Methodology, the description of relevant methods is too brief.
3. In section 5, the results were not fully discussed, and the internal correlation of the ten factors could be discussed in depth.
4. The titles of Figure 3 and Table 2 do not correspond.
5. “An example is the calculation of the ORI of the occupational accident factor "2.1.6 Electrical hazards/electrocution",2.1.6 should be changed to “2.1.13 Electrical installation/electrocution hazards”.
6. Table2 in Section 5 should be changed to table3.
7. There is something wrong in Table, for example, 3.0 Worker Training should be Worker Training Deficiencies.
Author Response
Please see attachments
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The article describes the literature analysis on accidents at work. Therefore, it should be treated as a review article and should be better written. Indeed, the authors conducted a broad literature review and rejected articles and research in master's and doctoral theses, but the documentation is modest.
The article’s aim is presented in Line 101: “The aim of this study is the quantitative analysis of data from existing studies, intending to crystalize the main accident hazards and factors in the construction sector in Greece by synthesizing the results of multiple similar studies.”
The purpose is clear, but it’s not a scientific problem. The problem is also that the authors did not clearly present the research despite the fact that they collected this data. The MSExcel spreadsheet is mentioned, but the reader needs to know how the study was done.
Also, the research results need to be sufficiently presented. It's not legible. The discussion that took place was very modest.
The authors in their article presented conclusions, but they are apparent. They noted that the analysis concerns only one market (Greece) and gave the need for further research covering a broader region. This shows that the authors know that the study must be completed and may only interest Greek readers.
Author Response
Please see the attachement
Author Response File: Author Response.pdf
Reviewer 3 Report
This is a very well done scientific work that makes the meta analysis of the reasons for the occurrence of occupacional accidents in construction sector, based on published scientific articles.
some aspects to improve:
line 101 - Construct the sentence to better explain the context: these three factors are significantly associated with the occurrence of accidents among construction workers or their non-occurrence?
line 124 - The first time an acronym appears in the text it has to be written in full: Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA)
line 334 - This demo of ORI is presented in a confusing way. Build a clearer explanation
Tables 2 and 3 - ORI results must all have the same number of decimal places
Line 367 - This must be Table 3
Author Response
Please see attachment
Author Response File: Author Response.pdf
Reviewer 4 Report
The paper content is clear and does not give rise to questions.
Minor comment: Line 239 will = with
Author Response
Please see attachment
Author Response File: Author Response.pdf
Reviewer 5 Report
The paper addresses a meta-analysis of Greek construction industry studies in terms of accident contributing factors; however, some issues have been found as follows:
1. Line 125. To conduct the systematic literature review (SLR), the authors mention that they considered 226 sources, but it is necessary to specify what kind of source they used. Papers, reports, or another bibliographical source? Please include a brief explanation about such sources.
2. Lines 127 and 128. The authors conducted a statistical analysis of the findings of the SLR and content analysis based on specific inclusion criteria from which (25) studies were selected for meta-analysis. However, they do not include any detail about the “specific inclusion criteria” mentioned. It is crucial to know, for example, but not limited to, which parameters were considered and why, are those criteria coming from the literature or they were arbitrarily selected by the authors? Later, in section 3.3., the authors list the inclusion criteria, but they do not include any explanation about why they chose such criteria instead of others.
3. Lines 155 to 162. The authors mention a set of keywords; however, they do not include any solid reason to choose those keywords. Did the authors use a Monthly Search Volume (MSV) analysis or similar? It is recommended to add a solid explanation about why the authors chose the keywords listed in lines 155 to 162.
4. Line 163. The authors conducted a search in the Scopus (Elsevier) bibliographic database. Why did the authors use the Scopus database only? What about the Web of Science database, which includes journals with even better quartiles and impact factors? It is recommended to explain why the authors decided to use the Scopus database only and not other databases such as the Web of Science.
5. Lines 165 to 169. The authors found relevant studies coming from a set of journals (Automation in Construction, Safety Science, etc.). To better show the studies found and their corresponding journals, it is recommended to run a co-occurrence analysis by using some bibliometric softwares such as VOSViewer or Atlas.ti and include the outputs in the manuscript.
6. Line 191. Figure 1 does not show clearly how the authors jumped from 262 at the end of the screening phase to 165 sources, in comparison with the explanation given previously in lines 177 to 189. For example, in line 179 the authors mention 156 sources, but this number does not appear in the flowchart. It is important to be consistent throughout the manuscript. Therefore, it is recommended to match the numbers shown in lines 177 to 189 with the numbers shown in the flowchart. Otherwise, it is hard to follow the thread where the authors try to explain the PRISMA method applied to the search criteria they conducted.
7. Line 206 to 218. As previously mentioned, the authors did not include any explanation about why they picked the inclusion criteria and not others. To eliminate any potential bias or arbitrariness in the selection of the inclusion criteria, it is highly recommended to add a full explanation for each criterion including bibliographical citations. For example, for the first criterion:
Research focusing on factors that cause constriction accidents: according to AUTHOR 1 [XX] and AUTHOR 2 [YY], factors that cause construction accidents are considered relevant because of XXXXXXXXX.
And so on for the rest of the inclusion criteria.
8. Lines 252 to 255. Similar to what was observed about the solid reasons to select the inclusion criteria, here there is no explanations about the main five categories defined by the authors. As previously recommended, it is suggested to add a full explanation to support the selection chosen. Some procedures to support selection processes can be conducted by using either bibliographical analyses (as recommended in the previous point) or by running bibliometric analyses (co-occurrence, cross-referential analyses, etc.). The point is that the selection of the five categories has to be very well founded because the soundness of the rest of the analyses depends on how much strong the selection of the categories was.
9. Line 314. “Meta-analysis of data using the Overall Ranking Index (ORI)”. The authors begin emphasizing that the choice of the appropriate statistical model in a meta-analysis depends on the type of survey data included in it. However, they do not include bibliographical sources that utilize other methods to rank factors in the construction industry (fuzzy analysis, risk matrices, and of course ORI), and then, after citing other works they may conclude that the ORI is appropriate for their study. In this sense, it is recommended to write a short bibliographical review (no more than 1 or 2 paragraphs) mentioning other methods and then to wrap up such review stating why ORI is the best for the present study. The authors may take a look to the following papers which address different methods to rank factors in the construction industry (ordered by year, the newest first):
- Sharma, S., & Goyal, P. K. (2022). Fuzzy assessment of the risk factors causing cost overrun in construction industry. Evolutionary Intelligence, 15, 2269–2281.
- Chan, A. P., Nwaogu, J. M., & Naslund, J. A. (2020). Mental ill-health risk factors in the construction industry: systematic review. Journal of Construction Engineering and Management, 146(3), 04020004.
- Forcael, E., Morales, H., Agdas, D., Rodríguez, C., and León, C. (2018). “Risk Identification in the Chilean Tunneling Industry”. Engineering Management Journal, 30(3), 203-215.
- Wu, Z., Nisar, T., Kapletia, D. and Prabhakar, G. (2017), Risk factors for project success in the Chinese construction industry", Journal of Manufacturing Technology Management, 28(7), 850-866.
- Abbasi, G. Y., Abdel-Jaber, M. S., & Abu-Khadejeh, A. (2005). Risk analysis for the major factors affecting the construction industry in Jordan. Emirates Journal for Engineering Research, 10(1), 41-47.
Among others.
10. Lines 334 to 342. Table 2 is the core of the presented research; however, the example used to explain Table 2 is not clear enough, beginning with the number and name of the factor “2.1.6 Electrical hazards/electrocution”, whose name is later changed in the same paragraph (“2.1.6 Hazards from Electrical Installations/Electrocution”), but finally, it seems that the right name in Table 2 is “2.1.13 Electrical installation/electrocution hazards” because 2.1.6 corresponds to “Falling or slipping”. It is recommended to improve the explanation of Table 2 and modify the misspelled texts.
11. Lines 344 to 367. After running a complete analysis to rank the risk factors based on ORI, the section “Results and Discussion” was so weak; just two paragraphs and one table devoted to discussion is insufficient indeed. For example, based on the top ten factors found, a deeper analysis may be conducted as a comparison with previous studies for each of such factors, along with a comprehensive discussion in terms of lessons learned or recommendations for practitioners and so on. In summary, it is recommended to expand this section as much as the top ten factors found can be properly discussed.
General minor observations. Throughout the manuscript the words “Figure XX” and “Table YY” are not with uppercase (first letter). Some typos are also found in the paper (e.g. Excell is misspelled in line 333 or in the paragraph from lines 334 to 341: no.2 instead of No. 2 or No.10=1 instead of N10=1 with 10 as a subscript, etc.). Please check and correct.
Author Response
Please see attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have given good answers to the review comments and have revised the paper well.
Reviewer 2 Report
Despite the article's earlier rejection, the authors improved it significantly. In this form, the article is suitable for publication. I hope the authors also noticed that after corrections, it is better.
Reviewer 5 Report
Dear authors, this reviewer thanks for taking into account all his/her comments and observations, which were properly answered in the amended version of the manuscript.
