Incentive Schemes Increase Risky Behavior in a Safety-Critical Working Task: An Experimental Comparison in a Simulated High-Reliability Organization

Round 1

Reviewer 1 Report

I think the work should be published in some form. It can be improved substantially. 

1) Throughout the paper your in text citations/references did not converge. There are "errors" throughout where a reference of some kind should be visible.

2) the experimental design needs much more clarity for the reader, perhaps a figure or table or both to thoroughly communicate the recruitment process, the phases of the experimental procedure, the length of time for the total experiment and each phase, when and what measurements were taken, the contents of the training, the context in which the simulation/experiment took place, etc. etc. Although I think Table 3 and the corresponding text in section 2.3 is interesting and somewhat important but it depicts more about the stages within the simulated environment and doesn't provide enough detail to demonstrate that the experiment displayed internal validity--how did you isolate the effect of your independent variable on your dependent variable. This is a must for the reader to fully comprehend what and how you did this research and have confidence in the results. 

3) You measured for control variables but then did a single factor ANOVA? this could be made more clear. Why did you or didn't you include some or all of the control variables in the the ANOVA? The interaction terms may bolster your results to be able to explain what demographic features make a difference, if any, in the outcomes.

4) Figure where you communicate the results of the ANOVA (means with differences in significance by pairwise comparisons) needs some tweaking. Perhaps remove the pathways in the figure and use a footnote to denote which pairwise comparisons were significant. 

I am willing to read it again and provide additional thoughts after a revision.   

Author Response

Thank you very much for your comments and suggestions. In the attached document you will find all our adjustments in commented form.

Author Response File: Author Response.pdf

Reviewer 2 Report

Thank you for the opportunity to review this manuscript. Overall, I thought the study question was interesting and the design was creative and well done. I liked that the simulation was a start-up process, because this is a risky procedure in many manufacturing organizations. I appreciate that the authors sought to understand the role of incentives in safety - this is an important topic that is challenging to study. My primary concern is that the methods are confusing at times (particularly the description of the experimental procedure) and as the paper is currently written, could not be replicated. I think the authors should revise the methods to make the procedure clearer. I also struggled to understand the incentive contingencies and I think this could be presented in a table to take it clearer. I also think the experimental task could be explained in a table or by linking to a video that shows the program. The methods would benefit from more headings to organize the content and help the reader locate important procedure information. 

I found the introduction to be a little long and it included some information that wasn't entirely relevant to the study. I would suggest that they authors reduce the length of the manuscript and only describe research that is most relevant to the present study.

Finally, the paper could use a very thorough edit.

Author Response

Thank you very much for your comments and suggestions. In the attached document you will find all our adjustments in commented form.

Author Response File: Author Response.pdf

Reviewer 3 Report

Thank you very much for letting me read this paper. Personally, I think this paper needs further improvement. For example, there are too many paragraphs in the introduction part, and the expression is not very clear. The research results are too few, and the data analysis is not enough to support the hypothesis test. In short, the expression and logical structure of the whole paper give me a very bad feeling. thank you.

Author Response

Thank you very much for your comments and suggestions. In the attached document you will find all our adjustments in commented form.

Author Response File: Author Response.pdf

Reviewer 4 Report

It must have been a difficult situation, but I would like to thank the authors for their hard work in recruiting subjects and conducting the experiment over the course of 48 weeks. There are some studies for incentives and productivity, but few studies of risky behavior. In particular, since there are very few actual experimental cases like this manuscript, I think it is a very valuable manuscript, and it is expected to be used as a good reference material for other studies in the future. Since there are parts that need to be checked, I think that it will be a better manuscript if it is checked and supplemented.

1.  “2.4 control variables: ~~”

Age, gender, study duration,.... were mentioned, but the results for these variables were not actually presented. And I ended up saying “but this was not found to be necessary” on “line 404”.

Please present the results of the control variables and explain exactly why “but this was not found to be necessary”.

→ If the reason cannot be precisely explained, the results should be reanalyzed and presented again using a statistical method other than the used statistical method, ANOVA.

2.  line 400-402

There is no detailed explanation or rationale for deviations of the normal distribution. Please provide a rationale or add a detailed explanation.

3. Table 4

A significant correlation was found between “self-interest” and “work-safety tension”, but only mentioned in the results and not explained in the discussion. Add relevant content to your discussion.

4. Table 2

Add references to "Knowledge of unsafe procedures (0-5)" and "Knowledge of safe procedures (0-7)".

"Example items" are too simplistic to understand. Please explain by adding more details to the method or appendix.

5. line 64-65 “Employees use ~~ behavior 10.”

What is “10”? Is it a reference number?

6. line 68

Introduction → introduction

7. line 399

ANOVA → ANOVA (analysis of variance). Is it right?

8. (Error! reference source not found..)

There are many errors like this in the manuscript. Please check.

Author Response

Thank you very much for your comments and suggestions. In the attached document you will find all our adjustments in commented form.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Looks good. Thank you for trying to adjust based on my comments. I still think  figure 2 could be improved visually. There still remains some type of glitch in your referencing system that I suspect will be resolved through pre-print editing. The text reads: "Fehler! Verweisquelle konnte nicht gefunden werden." in numerous places throughout. Kind regards and best of luck. 

Author Response

Looks good. Thank you for trying to adjust based on my comments.

• Thank you for appreciating our efforts and your contribution to improving manuscript quality.

 I still think figure 2 could be improved visually.

• Thanks to your comment, we noticed that the automatic numbering of the figures does not work, so we have 2 "Figures 2". Once the chart for the flow and also the bar chart. We have already been able to correct this error.
• If you are referring to the flowchart, we are very open to suggestions, but as things stand at the moment, we do not have any approach for improvement, or do not see any difficulties in perception at the moment.
• With regard to the bar chart, we are of course also open to suggestions for improvement, but are also bound by the standard of the professional association. For the optimization of Figure 2 (actually figure 3), we tried to find a better form of representation in the margin between the standards of our domain, demands of the readership and the comprehensibility of the figure. In our estimation, the possible alternatives (to our knowledge) for the visualization would only lead to a worse grasp of the central information (violation value by condition and result of the statistical analysis).

There still remains some type of glitch in your referencing system that I suspect will be resolved through pre-print editing. The text reads: "Fehler! Verweisquelle konnte nicht gefunden werden." in numerous places throughout. Kind regards and best of luck.

• The problems with referencing seem to be a technical challenge, which we are in exchange with the editor office of the journal to solve. So far, we have checked all referencing on our end and completed it as needed.

Reviewer 2 Report

I appreciate the additional information that could aid in replication of the manuscript. I think the writing could use more editing. There are a lot of awkwardly worded sentences, and sentences with with more words than necessary. I would suggest having the paper professionally edited or using an editing software like Grammarly to clean up the writing.

Author Response

I appreciate the additional information that could aid in replication of the manuscript. I think the writing could use more editing. There are a lot of awkwardly worded sentences, and sentences with with more words than necessary. I would suggest having the paper professionally edited or using an editing software like Grammarly to clean up the writing.

• We are very pleased that our efforts are in line with your expectations. In the manuscript, we have checked all sentences with 3 or more lines of length for shortening potential. In total, we were able to shorten 23 text passages. The manuscript was subjected to professional proof-reading of the journal.

Reviewer 3 Report

I have no problem.

Author Response

I have no problem.

• Thank you for appreciating our efforts and your contribution to improving manuscript quality.

Reviewer 4 Report

Thank you for your hard work to correct it. It seems that the quality of the paper has improved. But still some errors are checked.

1. (Fehler! Verweisquelle konnte nicht gefunden werden)

What is this? Please check again.

Author Response

Thank you for your hard work to correct it. It seems that the quality of the paper has improved. But still some errors are checked.

1. (Fehler! Verweisquelle konnte nicht gefunden werden)

What is this? Please check again.

• Thank you for appreciating our efforts and your contribution to improving manuscript quality. The problems with referencing seem to be a technical challenge, which we are in exchange with the editor office of the journal to solve. So far, we have checked all referencing on our end and completed it as needed.