Leadership at the Extreme: A Longitudinal Study of Transformational Leadership Style and Well-Being in Firefighters
Round 1
Reviewer 1 Report
If this paper moves forward toward publication, it will need to be carefully revised. The authors seem to be leaning on a very outdated leadership model that doesn't reflect literature development from the past 30 years.
The line by line edits are as follows:
Line 25 Where is this not the case?
Line 34 There are many studies on FF well-being. Build a justification for your work that situates this study among them instead of claiming
that those studies are not out there.
Line 48 "Fighting forest fires"
Also, I'm not sure what you are comparing this to. Are you suggesting that structural fire fighters who fight forest fires consider it
more demanding than structural fires or that wildland firefighters consider it more demanding than something else?
Line 50 It's odd to lump the socio-cultural stress with the physical environment. At a minimum, this list should be composed of the impact
(stated as a stress state) rather than an external environmental issue.
line 57 If you've made the argument that firefighting is stressful, and there are simply more fires, why would there be any curiosity that more
stress would occur?
Line 66 The notion of a singular leader doesn't work in firefighting.
Line 71 This document needs to be proofread.
Line 76 What do you mean "for one week?"
Line 77 This seems backwards.
Line 83 There are many studies like this. I don't see what's unique.
Line 86 edit.
Line 101 How would two months be considered "longitudinal?"
Line 120 It's probably time to update this type of questionnaire. Maybe that will be one of your findings. The dimensions
are not discrete. For example, if someone "conveys" something, it has to be a "behavior." How else would you
know it? It has to be observable.
Line 234 When your research design produces a finding that is this illogical, it's time to revisit the variables and scales.
Author Response
Q1. If this paper moves forward toward publication, it will need to be carefully revised. The authors seem to be leaning on a very outdated leadership model that doesn't reflect literature development from the past 30 years.
R1. It Is true that many new leadership theories have emerged over the years, especially within the positive psychology tradition. However, we have performed a careful literature review of the transformational leadership and wellbeing literatures. We also did so within the scope of the firefighting activity. Although we agree that there has been research on leadership and wellbeing in firefighters, we found it very difficult to find longitudinal studies (rather then cross-sectional) that established any empirical link (cross sectional or longitudinal) between transformational leadership, job demands and wellbeing in firefighters. Studies examining the relationship between the three in other occupational contexts were also challenging to find (although there are few and we cite them). So, could you please provide a few references for the studies you mention? That would be very helpful, and we will gladly incorporate them in our manuscript.
Q2. Line 25 Where is this not the case?
R2. We have replaced “In Europe” with “Currently”, to make It broader.
Q3. Line 34 There are many studies on FF well-being. Build a justification for your work that situates this study among them instead of claiming
that those studies are not out there.
R3. We have changed the focus of our argument. We acknowledge your comment. In the new version of the manuscript, instead of focusing on the absence of studies (which to your point was not a precise argument), we now state that: “wellbeing plays a critical role in fire operations by reducing the number of accidents and ensuring the sustainability of fire operations [3]. Hence, expanding current knowledge on the psychological factors that enable fire fighters’ wellbeing during forest fires is of particular importance to empower fire chiefs and civil protection departments with the capacity to build firefighters’ wellbeing amidst fire operations [5].” As we can see, we have changed the focused from “the lack of studies” towards “the need to conduct additional research” especially focusing what happens during fire operations.
Q4. Line 48 "Fighting forest fires"Also, I'm not sure what you are comparing this to. Are you suggesting that structural fire fighters who fight forest fires consider it
more demanding than structural fires or that wildland firefighters consider it more demanding than something else?
R4. We are not establishing any comparisons, nor we intended too. We also did not control for forest vs. wildland type of fire. To avoid confusion, in the revised version of the manuscript we have adopted the term “fire events”.
Q5. Line 50 It's odd to lump the socio-cultural stress with the physical environment. At a minimum, this list should be composed of the impact
(stated as a stress state) rather than an external environmental issue.
R5. In this line we are simply stating that fire events can cause de depletion of mental and physical resources. Socio-cultural stress could exacerbate this relationship, but we are not making any reference to it nor it contributes to our theorizing. Also, since the sample is 100% constituted by a single nationality, socio-cultural stress might not be a critical factor.
Q6. line 57 If you've made the argument that firefighting is stressful, and there are simply more fires, why would there be any curiosity that more
stress would occur?
R6. We are addressing the number of fire events from a workload perspective. The more fire events firefighters must address over a week, the more demanding their activity will be, and wellbeing will be at risk. Also, firefighting is a stressful occupation (Huynh et al., 2014) because firefighters must tackle stressful events like fires. Naturally, there are other stressful events to which firefighters must respond; and not all firefighting activity is stressful because not all distress calls are severe, there are long waiting times that can be monotonous, and there are plenty of opportunities for doing sports.
Q7. Line 66 The notion of a singular leader doesn't work in firefighting.
R7. We are not making such statement. The sentence to which you are referring too simply states that fire chiefs’ leadership behaviors have implications for what others will do. This is not denying the existence of shared leadership processes, as well as leader-follower or informatal leadership dynamics (as examples). Nonetheless, it goes without saying that in the specific case of our sample and in how the national firefighting system is organized, firefighters belonged to teams that acted in emergency situations, with priority for fires in rural areas (not-urban). These teams are made up of 5 members, one of whom is the head, and are “permanent” / stable, which means that there is no rotation in the leadership role.
Q8. Line 71 This document needs to be proofread.
L8. This has been taken care off. We have also rewritten the sentence: “Additionally, the leader uses inspirational motivation and/or individualized consideration to help members believe in their skills and abilities. By doing so, the leader promotes the development of firefighters’ self-efficacy beliefs that they will be able to face the most demanding situations with enthusiasm and dedication.”
From this point onwards, the comments and the lines you mention did not have an exact correspondence in the manuscript. We have tried to locate the problem and respond accordingly.
Q9. Line 76 What do you mean "for one week?"
R9. We believe this comment regards Line 82 (Hypothesis 2). We mean “seven consecutive days”. We have replaced “for one week” with “seven consecutive days”.
Q10. Line 77 This seems backwards.
R10. We could not identify which sentence was backwards. Since we have sent the manuscript for proofread, we hope this has been solved.
Q11. Line 83 There are many studies like this. I don't see what's unique.
R11. You will agree that most research is incremental and that there is an urgent need for replication studies that support previous findings in different contexts and conditions. Nonetheless, the uniqueness in our studies resides in the sample, national context, and the fact that the year in which our data was collected was one of the most devastating so far. We did not see this as important to speak to the uniqueness of the research, but we will gladly include it if you think it is.
Also, we have adjusted the final part of the introduction section to make the contribution of the manuscript clearer. As you may notice, longitudinal studies looking at how leadership contributes to the temporal relationship between fire events and wellbeing are not abundant. They are hard to implement, especially during the peak of the fire season: “We test these hypotheses through a longitudinal methodology to study the role that the frequency of fire events (as operation demands) , and fire chiefs’ transformational leadership (as operation resources) have in explaining how firefighters’ wellbeing changes over seven consecutive days. In so doing, this study goes beyond previous research because it considers the role of fire operations task conditions in the health impairment process associated with daily decreases in firefighters’ wellbeing. This study does so by using a longitudinal approach that helps clarifying the temporal relationship between fire events and firefighter wellbeing.”
Q12. Line 86 edit.
R12. Could not understand the “Edit” suggestion.
Q13. Line 101 How would two months be considered "longitudinal?"
R13. It is not the two months (July and August) that make the study Longitudinal, but rather the fact that our design includes the collection of the same variables on more then 3 occasions over time. See Roe, R. A. (2008). Time in applied psychology: The study of" what happens" rather than" what is.". European Psychologist, 13(1), 37.
Q14. Line 120 It's probably time to update this type of questionnaire. Maybe that will be one of your findings. The dimensions are not discrete. For example, if someone "conveys" something, it has to be a "behavior." How else would you know it? It has to be observable.
R14. We could not understand your point here. Please clarify.
Q15. Line 234 When your research design produces a finding that is this illogical, it's time to revisit the variables and scales.
L15. Variables and scales are always prone to improvement, as they are also sensitive to contexts and occupations. That said, we could not understand to which finding you are referring too when you say “illogical”. Could you please clarify?
Reviewer 2 Report
Thank you for the possibility to read this contribution. I was looking forward to this work. This study seeks to determine whether transformational leadership style has an impact on the well-being of firefighters. The occurrence of forest fires is considered as a stress factor. It is a very interesting and also very important field of research.
I will gladly go into some points more explicitly in the following. For this, I will address major but also minor issues.
Keywords
I would suggest to include also “firefighters” into the keywords
Introduction
When mentioning HRO in line 26 it would be appropriate to cite Weick and Sutcliffe (2007), Managing the Unexpected. It is THE reference. Source 1 can remain here, but in any case, should not stand alone on this term.
Examples for the operational demands in line 28 would be helpful for the readers.
In line 34 the insertion "in HRO such as" should be deleted. It is sufficient to write "wellbeing in fire operations", since this is also exactly what is involved. Moreover, the question regarding HROs is not answered in this paper either.
The logic of the following sentence in line 35 is not clear. What does this have to do with the previous sentence? Verbally, a chain of reasons is built up here, but its content cannot be followed. Why exactly are these examples excluded? What are the authors trying to get at here?
What are the limitations referred to in line 37? More clarification is needed here, please.
Starting at line 46, the frequency of forest fires is discussed. I consider this term to be inaccurate and actually not relevant for the present work. I consider the frequency of the operations in which they themselves participated and not the number of forest fires to be relevant as a burden for the emergency forces. I therefore doubt this construct in its entirety. The authors please specify this, if they can, and provide data that differentiates this.
A visualization of the theoretical model and an integration of the hypotheses into it would be very useful and desirable.
Starting at line 64, the transformational leader-member relationship is discussed and what is part of it. Among other things, richness of verbal communication and bi-directional feedback. However, I now wonder how all of this is possible in one mission. Here the radio must not be overloaded and it is extremely important to make sure that only the important information is given to the right people at the right time and that a good situation awareness is maintained in the mission. So how should what is described here work? What is the mechanism of action of transformational leadership here? More detailed explanations of the mechanism of action and also the timing are desirable and necessary.
The reference in hypothesis 2 to one week is not justified. Why is it assumed that well-being will decrease within one week if transformational leadership is low? Why is one week relevant? What happens in this one week? Isn't transformational leadership a construct that has a much more lasting effect? This fact needs to be explained in detail please and also the derivation of the hypothesis.
Unfortunately, the second part of the hypothesis is also not comprehensible to me. Why should the work engagement increase with forest fires? What is actually meant by forest fires? Does this mean a frequency or is it also sufficient if someone has been on a forest fire mission once? And why is an increase assumed with a high transformational leadership style? No further change would be conceivable, wouldn't it? The justification of this hypothesis needs to be supported by theory in more detail.
One last small note in this section. In line 87, what are leaders' characteristics? This indicates person-related variables. More in the area of personality. However, is this perhaps more about leadership style? Since this is a leadership style, these are not characteristics of the leader, but are behaviors of the leader.
Materials and Methods
Please specify the year in line 101.
Line 126: please write 5-point. Please repeat this at the other required points.
It is not clear how data were handled when participants said that they were not assigned to a rural fire operation on a day, but did fill in the questionnaires. Please explain this procedure.
Results
Please insert a table with all descriptive data at the beginning of the section results, including mean, SD and correlations for all study variables.
It would also be very interesting to see the mean values for all dependent variables at each time point to visualize any changes.
In line 218, I do not understand the reference to tables 3 and 4. Shouldn't these be tables 1 and 2? Also, it is very unusual to put tables in the appendix if they are important to answer the hypothesis. If this is not the case for Tables 1 and 2, why are they in the text and not in the appendix? I can't follow the structure, there is a clear confusion for me.
Why are the significant results from Table 2 not discussed and why are they not used to test Hypothesis 1(a)?
Regarding the analysis of the data, I would suggest to calculate growth curve models.
Discussion
The discussion turns out to be extremely short. It is desirable that now the individual hypotheses are discussed and the findings are put back into the previous literature. The literature work in the discussion is insufficient.
I am surprised that the duration of the shifts is not discussed, however, since a significant influence has been shown here.
Overall, I am very suspicious of the discussion and think it is inappropriate to suggest that transformational leadership has such a positive impact, which could hardly be shown here, not one hypothesis was supported. Moreover, I ask why the findings are presented in such a positive light, since the positive effects of transformational leadership on employees have already been shown many times. So why should it be assumed that they do not also apply in HROs? According to this, the results would rather be a replication of other studies just now in the fire service as well. There are numerous papers, not mentioned here, that relate leadership behavior to the JD-R model.
Unfortunately, the discussion is not sufficient for a scientific publication. More theory work is needed, a discussion of external and internal validity as well as limitations is missing.
I really appreciate the work that was put into the study. The fact that no changes can be seen over the short period of time is not unexpected. I would like to see a more reflective approach to the data and analyses and think that the paper will benefit greatly from this.
The paper needs a linguistic proofreading.
Author Response
Q1. Thank you for the possibility to read this contribution. I was looking forward to this work. This study seeks to determine whether transformational leadership style has an impact on the well-being of firefighters. The occurrence of forest fires is considered as a stress factor. It is a very interesting and also very important field of research.
I will gladly go into some points more explicitly in the following. For this, I will address major but also minor issues.
R1. Thank you for your positive comment.
Keywords
Q2. I would suggest to include also “firefighters” into the keywords
R2. Done.
Introduction
Q3. When mentioning HRO in line 26 it would be appropriate to cite Weick and Sutcliffe (2007), Managing the Unexpected. It is THE reference. Source 1 can remain here, but in any case, should not stand alone on this term.
R3. Done. However, you will notice that we have tried to abandon the term HRO and focused mainly on Firefighters.
Q4. Examples for the operational demands in line 28 would be helpful for the readers.
R4. Done.
Q5. In line 34 the insertion "in HRO such as" should be deleted. It is sufficient to write "wellbeing in fire operations", since this is also exactly what is involved. Moreover, the question regarding HROs is not answered in this paper either.
R5. Done.
Q6. The logic of the following sentence in line 35 is not clear. What does this have to do with the previous sentence? Verbally, a chain of reasons is built up here, but its content cannot be followed. Why exactly are these examples excluded? What are the authors trying to get at here?
R6. Thank you for this comment. As you may have the opportunity to see, we have made substantial modifications in the paper, including the structure and grammar of the introduction section. We believe this issue is now fixed.
Q7. What are the limitations referred to in line 37? More clarification is needed here, please.
R7. We think this is now clear.
Q8. Starting at line 46, the frequency of forest fires is discussed. I consider this term to be inaccurate and actually not relevant for the present work. I consider the frequency of the operations in which they themselves participated and not the number of forest fires to be relevant as a burden for the emergency forces. I therefore doubt this construct in its entirety. The authors please specify this, if they can, and provide data that differentiates this.
R8. Thank you for bringing this concern. We thought about it and the fact is that the firefighters that participated in this study were mostly allocated to forest fires during this time. We agree that other events may have happened, although unfortunately we could not account for them. We have addressed this issue in the limitation.
Q9. A visualization of the theoretical model and an integration of the hypotheses into it would be very useful and desirable.
R9. Thank you for this suggestion. A new figure has been included that should address this concern.
Q10. Starting at line 64, the transformational leader-member relationship is discussed and what is part of it. Among other things, richness of verbal communication and bi-directional feedback. However, I now wonder how all of this is possible in one mission. Here the radio must not be overloaded and it is extremely important to make sure that only the important information is given to the right people at the right time and that a good situation awareness is maintained in the mission. So how should what is described here work? What is the mechanism of action of transformational leadership here? More detailed explanations of the mechanism of action and also the timing are desirable and necessary.
R10. Thank you for bringing this into discussion. We agree that most communication during missions is minimized to operational communications, with little opportunities for a more positive way of leading. It could even be that during operations, transactional leadership rather then transformational leadership is more active. However, briefings and debriefings which are part of the operational routine should be good occasions for transformational leadership to happen. Sadly, we could not control for when did fire chiefs engage in transformational leadership. We have tried to address this issue in the discussion.
Q11. The reference in hypothesis 2 to one week is not justified. Why is it assumed that well-being will decrease within one week if transformational leadership is low? Why is one week relevant? What happens in this one week? Isn't transformational leadership a construct that has a much more lasting effect? This fact needs to be explained in detail please and also the derivation of the hypothesis.
R11. The reason for a one-week data collection was a practical one: we could only gather data for one week. Additionally, since we regarded transformational leadership as a dispositional variable (i.e., attribute like), we assume that leaders perceived as transformational will be consistent in their displays of transformational leadership and therefore there was no point in measuring transformational leadership on a daily basis.
Q12. Unfortunately, the second part of the hypothesis is also not comprehensible to me. Why should the work engagement increase with forest fires? What is actually meant by forest fires? Does this mean a frequency or is it also sufficient if someone has been on a forest fire mission once? And why is an increase assumed with a high transformational leadership style? No further change would be conceivable, wouldn't it? The justification of this hypothesis needs to be supported by theory in more detail.
R12. This was a typo, which is now corrected and clarified.
Q13. One last small note in this section. In line 87, what are leaders' characteristics? This indicates person-related variables. More in personality. However, is this perhaps more about leadership style? Since this is a leadership style, these are not characteristics of the leader, but are behaviors of the leader.
R13. We have clarified this in the revised version of the manuscript.
Materials and Methods
Q14. Please specify the year in line 101.
R14. Done
Q15. Line 126: please write 5-point. Please repeat this at the other required points.
R15. Done.
Q16. It is not clear how data were handled when participants said that they were not assigned to a rural fire operation on a day, but did fill in the questionnaires. Please explain this procedure.
R16. The frequency of fire events over 7 days was averaged and introduced in the RCM model as an input. The average was based on the frequencies of Yes there was a fire I attended VS. No there wasn’t a fire I attended. We have tried to make this clearer in the new manuscript.
Results
Q17. Please insert a table with all descriptive data at the beginning of the section results, including mean, SD and correlations for all study variables.
R17. We have included a new table with all the means and standard deviations for all the research variables. Regarding the repeated measures correlation table, since only work engagement and emotional exhaustion were significantly correlated, and to avoid overloading the manuscript with an additional table, we believe that we are still providing all the necessary information without compromising rigor nor readability.
Q18. It would also be very interesting to see the mean values for all dependent variables at each time point to visualize any changes.
R18. Done
Q19. In line 218, I do not understand the reference to tables 3 and 4. Shouldn't these be tables 1 and 2? Also, it is very unusual to put tables in the appendix if they are important to answer the hypothesis. If this is not the case for Tables 1 and 2, why are they in the text and not in the appendix? I can't follow the structure, there is a clear confusion for me.
R19. It was a typo. We have corrected it. Thank you.
Q20. Why are the significant results from Table 2 not discussed and why are they not used to test Hypothesis 1(a)?
R20. Thank you for this comment. We have tried to integrate it.
Q21. Regarding the analysis of the data, I would suggest to calculate growth curve models.
R21. Thank you. We also though of GCM models in the beginning of the research, but we have to many parameters to estimate given the sample size and structural equation models are unforgiven regarding the ration of participants to parameters. Therefore, we decided to use Random Coefficient Models, which provide similar results to GCM and are less sensitive to sample size. See for example: Ployhart, R. E., & Vandenberg, R. J. (2010). Longitudinal research: The theory, design, and analysis of change. Journal of management, 36(1), 94-120.: or Ployhart, R. E., & Ward, A. K. (2011). The “quick start guide” for conducting and publishing longitudinal research. Journal of Business and Psychology, 26(4), 413-422.
Discussion
Q22. The discussion turns out to be extremely short. It is desirable that now the individual hypotheses are discussed and the findings are put back into the previous literature. The literature work in the discussion is insufficient.
R22. We have worked substantially on the discussion, following your suggestion.
Q23. I am surprised that the duration of the shifts is not discussed, however, since a significant influence has been shown here.
R23. We have addressed this in the new version of the manuscript.
Q24. Overall, I am very suspicious of the discussion and think it is inappropriate to suggest that transformational leadership has such a positive impact, which could hardly be shown here, not one hypothesis was supported. Moreover, I ask why the findings are presented in such a positive light, since the positive effects of transformational leadership on employees have already been shown many times. So why should it be assumed that they do not also apply in HROs? According to this, the results would rather be a replication of other studies just now in the fire service as well. There are numerous papers, not mentioned here, that relate leadership behavior to the JD-R model. Unfortunately, the discussion is not sufficient for a scientific publication. More theory work is needed, a discussion of external and internal validity as well as limitations is missing.
R24. We sincerely hope that you find your concerns addressed by the new version of the paper.
I really appreciate the work that was put into the study. The fact that no changes can be seen over the short period of time is not unexpected. I would like to see a more reflective approach to the data and analyses and think that the paper will benefit greatly from this.
The paper needs a linguistic proofreading.
Reviewer 3 Report
Thank you for the opportunity to review an interesting and well presented manuscript. Overall there are some interesting findings presented. The following comments are provided as a means to enhance the manuscript further.
General Comments:
The biggest question I have with the manuscript is what actually occurred during the 7 days of the assessment period. I understand that this period was in a period when forest fires are most prominent but did these occur during that week? How much stress was placed on the participants during this time?
Can the authors also expand on the leadership behaviour. Is it possible to know that the leader behaviour remained constant during that time? While the data collection may suggest that the leaders preferred a transformational approach, was this actually implemented and adhered to during highly stressful situations. How consistent was the leadership.
The discussion and conclusion is very short. I would like to see the authors expand on the main components and add more depth to the analysis. What is the implications of this knowledge? How can this information be used as an education for the leaders? What recommendations can be made?
Specific comments:
L37 - please check the reference format
L57 - change to 'the frequency of forest fire occurrences is positively related to emotional exhaustion'
L106 - change 'annonin' to 'anonymous'
Table 2 - change 'firest fires' to 'forest fires'
Author Response
Q1. Thank you for the opportunity to review an interesting and well presented manuscript. Overall there are some interesting findings presented. The following comments are provided as a means to enhance the manuscript further.
R1. Thank you for the positive comment.
Q2. The biggest question I have with the manuscript is what actually occurred during the 7 days of the assessment period. I understand that this period was in a period when forest fires are most prominent but did these occur during that week? How much stress was placed on the participants during this time?
R2. We cannot tell exactly what was happening during the 7 days period, except that the frequency of fire events tells us how many fire events each participant attended during the week of data collection. However, we do not know for sure the level of stress that was experienced by our participants because we did not collect that data (either quantitatively or qualitatively). We have now addressed this in the discussion section.
Q3. Can the authors also expand on the leadership behaviour. Is it possible to know that the leader behaviour remained constant during that time? While the data collection may suggest that the leaders preferred a transformational approach, was this actually implemented and adhered to during highly stressful situations. How consistent was the leadership.
R3. This is a very interesting question. Thank you. In fact, we cannot know. This is something we also included in the new version of the manuscript in the discussion section, by alluding to the fact that event characteristics might change the relationship between variables or how variables change over time. We have studied transformational leadership from an attribute perspective, hence asking participants about their fire chiefs transformational leadership style. By doing so we expect some consistency in leaders’ behaviors, although we are aware that contexts and situations can favor other leadership behaviors such as empowering and transactional leadership.
Q4. The discussion and conclusion is very short. I would like to see the authors expand on the main components and add more depth to the analysis. What is the implications of this knowledge? How can this information be used as an education for the leaders? What recommendations can be made?
R4. We have expanded the discussion section, mostly focusing on research limitations and how these could be addressed in future studies.
Specific comments:
Q5. L37 - please check the reference format
R5. Done
Q6. L57 - change to 'the frequency of forest fire occurrences is positively related to emotional exhaustion'
R6. Done
Q7. L106 - change 'annonin' to 'anonymous'
R7. Done
Q8. Table 2 - change 'firest fires' to 'forest fires'
R8. Done
Round 2
Reviewer 2 Report
I read the revised paper very carefully. I will now return to the points that require further clarification. If necessary, there are new points to discuss.
Introduction
Q3. When mentioning HRO in line 26 it would be appropriate to cite Weick and Sutcliffe (2007), Managing the Unexpected. It is THE reference. Source 1 can remain here, but in any case, should not stand alone on this term.
R3. Done. However, you will notice that we have tried to abandon the term HRO and focused mainly on Firefighters.
Q: I must say that I am now irritated to the highest degree about this revision. Why was the term HRO deleted and then Weick and Sutcliffe quoted? None of this makes any sense anymore. If you now want to say that the missions have a negative impact on well-being, then it would be very good if you would cite a suitable scientific source that has studied exactly that. The meaning of the sentences has completely changed in this revision.
Q8. Starting at line 46, the frequency of forest fires is discussed. I consider this term to be inaccurate and actually not relevant for the present work. I consider the frequency of the operations in which they themselves participated and not the number of forest fires to be relevant as a burden for the emergency forces. I therefore doubt this construct in its entirety. The authors please specify this, if they can, and provide data that differentiates this.
R8. Thank you for bringing this concern. We thought about it and the fact is that the firefighters that participated in this study were mostly allocated to forest fires during this time. We agree that other events may have happened, although unfortunately we could not account for them. We have addressed this issue in the limitation.
Q: This point, concerning hypothesis 1, is very important because the data are essential to answer the hypotheses. It is good that this topic is now addressed in the limitations, even if it could be more detailed there. Nevertheless, I wonder what data was collected. The integration of Table 1 is very important and good, there I see now the mean values to the missions in the one week. However, it says in the methods section that 1 or 2 was used to ask if there was an operation or not. This is a dichotomous variable and you are not allowed to calculate a mean value. Here it is important to show how many of the 115 people on the days indicated they had experienced a firefighting operation. Then one would see what is being talked about at all.
Since the people were supposed to indicate whether or not they were assigned to a fire call that day, the data would be there to speak of the frequency of operations in which they were involved, wouldn't it? Then the derivation of hypothesis 1 could be modified accordingly and then the data would also fit to answer the hypothesis. I strongly recommend revising the derivation of the hypothesis.
Q9. A visualization of the theoretical model and an integration of the hypotheses into it would be very useful and desirable.
R9. Thank you for this suggestion. A new figure has been included that should address this concern.
Q: Nice to see a Figure. Unfortunately, however, the figure does not reflect a hypothesis model. It is not possible to see graphically what the hypotheses look like. It seems like a mixture of the hypotheses and the procedure of the study. I suggest creating the figure so that the hypotheses are visible. Feel free to look at other published papers on how correlation hypotheses are represented with different directions and moderators.
Q11. The reference in hypothesis 2 to one week is not justified. Why is it assumed that well-being will decrease within one week if transformational leadership is low? Why is one week relevant? What happens in this one week? Isn't transformational leadership a construct that has a much more lasting effect? This fact needs to be explained in detail please and also the derivation of the hypothesis.
R11. The reason for a one-week data collection was a practical one: we could only gather data for one week. Additionally, since we regarded transformational leadership as a dispositional variable (i.e., attribute like), we assume that leaders perceived as transformational will be consistent in their displays of transformational leadership and therefore there was no point in measuring transformational leadership on a daily basis.
Q: This is a really bad argument and not compatible with good research. Actually, the results are thus rather random. Who knows what happens to well-being after two weeks and why? No one, if one does not think about the time period in advance. This must be included in the discussion as a major limitation. It is also a very short period of time for leadership to have an effect.
Results
Q17. Please insert a table with all descriptive data at the beginning of the section results, including mean, SD and correlations for all study variables.
R17. We have included a new table with all the means and standard deviations for all the research variables. Regarding the repeated measures correlation table, since only work engagement and emotional exhaustion were significantly correlated, and to avoid overloading the manuscript with an additional table, we believe that we are still providing all the necessary information without compromising rigor nor readability.
Q: Please see my comment above and do not calculate a mean value for a dichotomous variable.
Discussion
Discussion of the unsupported hypotheses is lacking. What does it say that these had to be rejected? Why might it be that the hypotheses could not be supported? What are the reasons for this? A discussion of each hypothesis is missing. I ask to integrate these with literature deposited.
Author Response
Dear Reviewer 2,
Thank you for taking time for another in-depth revision of our manuscript. We provide point-by-point responses to each of your concerns. You will see how we have improved the manuscript following your suggestions.
Introduction
Q3. When mentioning HRO in line 26 it would be appropriate to cite Weick and Sutcliffe (2007), Managing the Unexpected. It is THE reference. Source 1 can remain here, but in any case, should not stand alone on this term.
R3. Done. However, you will notice that we have tried to abandon the term HRO and focused mainly on Firefighters.
Q3.1: I must say that I am now irritated to the highest degree about this revision. Why was the term HRO deleted and then Weick and Sutcliffe quoted? None of this makes any sense anymore. If you now want to say that the missions have a negative impact on well-being, then it would be very good if you would cite a suitable scientific source that has studied exactly that. The meaning of the sentences has completely changed in this revision.
R3.1. There were three Reviewers assigned to our manuscript. The removal of HRO and the inclusion of Weick and Sutcliffe was done following the recommendations from two other reviewers, to include relevant research and increase the precision of our theorizing having context into consideration. As you may know, HRO organizations can include air traffic control teams, and nuclear power plant teams, as examples, and neither of those is as exposed to risky activities as Firefighters and the characteristics of the job also different significantly (Hällgren et al., 2018). In the new version of the manuscript, we have tried to recover HRO by adding a final sentence to our discussion, where we speak to the importance of our findings to other HRO occupations such as air-traffic control and nuclear power plants.
-----
Q8. Starting at line 46, the frequency of forest fires is discussed. I consider this term to be inaccurate and actually not relevant for the present work. I consider the frequency of the operations in which they themselves participated and not the number of forest fires to be relevant as a burden for the emergency forces. I therefore doubt this construct in its entirety. The authors please specify this, if they can, and provide data that differentiates this.
R8. Thank you for bringing this concern. We thought about it and the fact is that the firefighters that participated in this study were mostly allocated to forest fires during this time. We agree that other events may have happened, although unfortunately we could not account for them. We have addressed this issue in the limitation.
R8.1. In have further addressed this in the discussion section. Table 1 has been updated to include the frequencies of fire events. We also modified H1 and H2. H1 is about the occurrence (occurred vs not occurred) over time, not the frequency. In the previous version of the manuscript, it was said that the predictor variable was the frequency of events. This was a typo and has now been corrected across the manuscript.
----
Q9. A visualization of the theoretical model and an integration of the hypotheses into it would be very useful and desirable.
R9. Thank you for this suggestion. A new figure has been included that should address this concern.
R9.1. The Figure has now been revised to include the Hypotheses.
----
Q11. The reference in hypothesis 2 to one week is not justified. Why is it assumed that well-being will decrease within one week if transformational leadership is low? Why is one week relevant? What happens in this one week? Isn't transformational leadership a construct that has a much more lasting effect? This fact needs to be explained in detail please and also the derivation of the hypothesis.
R11. The reason for a one-week data collection was a practical one: we could only gather data for one week. Additionally, since we regarded transformational leadership as an attribute variable, we assume that leaders perceived as transformational will be consistent in their displays of transformational leadership and therefore there was no point in measuring transformational leadership on a daily basis.
Q: This is a really bad argument and not compatible with good research. Actually, the results are thus rather random. Who knows what happens to well-being after two weeks and why? No one, if one does not think about the time period in advance. This must be included in the discussion as a major limitation. It is also a very short period of time for leadership to have an effect.
R1.11 A comment in the discussion section was added, mentioning the 1-week timeline as a limitation, and suggesting other alternatives. In the methods section it was already said that data collection was collected between July and August to maximize chances of collecting data during fire occurrences. We agree that a longer data collection, or even multiple daily collections to allow a diary study would probably render additional insight. However, to the best of our knowledge, there is no theory that can recommend on the optimal time scale to study emotional exhaustion and work engagement. Additionally, regarding the honesty of saying that the 1-week window was a practical decision, this reflects the challenges inherent to field research in extreme settings and therefore is not a commonsense argument (e.g., Bell, S. T., Fisher, D. M., Brown, S. G., & Mann, K. E. (2018). An approach for conducting actionable research with extreme teams. Journal of Management, 44(7), 2740-2765).
Results
Q17. Please insert a table with all descriptive data at the beginning of the section results, including mean, SD and correlations for all study variables.
R17. We have included a new table with all the means and standard deviations for all the research variables. Regarding the repeated measures correlation table, since only work engagement and emotional exhaustion were significantly correlated, and to avoid overloading the manuscript with an additional table, we believe that we are still providing all the necessary information without compromising rigor nor readability.
Q: Please see my comment above and do not calculate a mean value for a dichotomous variable.
R17.1 Done
Discussion
Q: Discussion of the unsupported hypotheses is lacking. What does it say that these had to be rejected? Why might it be that the hypotheses could not be supported? What are the reasons for this? A discussion of each hypothesis is missing. I ask to integrate these with literature deposited.
R18.1 We have revised the discussion section to ensure that we included additional thoughts on the unsupported hypotheses.
