Telework and Face-to-Face Work during COVID-19 Confinement: The Predictive Factors of Work-Related Stress from a Holistic Point of View

Round 1

Reviewer 1 Report

Dear authors, I am very pleased to read your paper.
It is well written and I think that, after some improvements, it will deserve the publication on the IJERPH.

Here you find some comments on your manuscript.

1. Please, clarify what "holistic" means from the very beginning.
2. Lines 44-51. This risks to be false. Dozens of papers refer to contexts in which both face to face work and telework are considered together. I would give less emphasis to this sentence.
3. Please state your hypotheses as separate from the text and linking them to the literature they are based on.
4. In the literature, I do think that you should give more emphasis on the fact that teleworkers live differently and react differently to stress because it is possible to individuate among them different "profiles" of them. As an example, I suggest this contribute from our journal: https://www.mdpi.com/1660-4601/18/22/12095
   And this contribute from another MDPI journal: https://www.mdpi.com/2071-1050/12/13/5332/
5. I would deepen the argumentation about the family care, more explicitly addressing the issue of children care (or others' care in general) with COVID-19 related literature (you mainly use "older" literature but this point is very COVID-19 related). I suggest to further argue about this aspect, and I point you to two papers that address it: https://journals.lww.com/joem/Fulltext/2021/03000/Impacts_of_Working_From_Home_During_COVID_19.2.aspx and https://econtent.hogrefe.com/doi/full/10.1024/2673-8627/a000015 
6. I would place less importance on the "holistic" definition you use. I would try to use the words "contextual, organizational, and individual" more. I think "holistic" misses the point, although then explained in the paper.
7. I think "dependent" is not the best word to indicate that you indicate. May you use another term? It seems "to be an employer and having employees" (dependent work).
8. Please, make the tables more readable and consider to insert figures to summarize the tested hypotheses or the results.
9. The results between telecommuters and non-telecommuters are broadly similar. This dismantles a bit the power of the paper, which bases part of its "originality" on the consideration of the two groups.
   What do you think? How can this become a resource for this paper?
10. Can you provide a little conclusion to the paper? It may help it to have a better "end" for the manuscript.

I hope to have indicated some directions for improvement.
Keep up the good work and congratulations.
Best regards

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present the results of a survey about the factors which predict work-related stress in face-to-face workers and teleworkers during the confinement due to the Covid-19 pandemic.

Methods and statistical content appear to be broadly reasonable. The main problem of this research lies in its relatively small convenience sample. It is not probabilistic and does not achieve representative survey populations, therefore it's hard to tell whether the obtained results could be generalized in the wider population. Moreover, the authors do not acknowledge this strong limitation of their research, which could also explain why some of their results differ from existing literature. In my opinion, the authors should include this limitation in the final section and tone down their conclusions accordingly.

Lastly, in order to be acceptable, the manuscript strongly needs to be checked by a native proofreader

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors present the results of an online survey of workers in the first months of the pandemic COVID. They conducted a cross-sectional study and examined how work-related stress was related to economic threat, family context, working hours, and safety and organisational resources. In addition, they investigated whether gender or income, among other factors, moderated the above relationships.
The introduction is well written.
Objective. Unclear to me "having dependents": did the authors mean taking care of a dependent person?
Hypotheses. The authors list 10 main hypotheses plus subsidiary hypotheses, I would say too much for a reader. I strongly recommend the authors to revise this paragraph. They mainly reasoned around three associations, each moderated by variables such as gender, care of a dependent person or income level: Economic Threat and Stress (H 1 to 3), Work-Family Conflict and Stress (H 3 to 5), and Rumination and Stress (H 9 and 10), as well as hypotheses related to each subsample (H 6 to 8; teleworkers versus face-to-face workers).
Participants. Were they recruited using a chain sampling technique? Did the authors pay attention to which participants worked for the same organisation? In other words, did they pay attention to nesting workers in their organisational structures? If not, why not? (Already partially acknowledged in Limits )
Variables. These are self-reported variables; however, sometimes their names suggest otherwise. For example, the degree of responsibility for the organisation ranges from "No" to "Very High", whereas the reader would expect the degree of responsibility to be determined on the basis of more objective job classification systems; similarly, Worked hours is a variable that is qualitative rather than quantitative, i.e., more or less than before COVID. Job stress is also assessed in its emotional components, while areas such as demands, support and control, among others, are completely disregarded. Why?
Data analysis and Results. It is beneficial for the reader to know that the Hayes et al. macro process allows for moderation analysis to be conducted in SPSS.
Hypotheses and Results sections should be consistent to each other.
Table 1. t-test comparisons should be provided.
Table 2. not required; results (along with a chi-square statistic) could be provided in the text.
Table 3. a typo ('genre').
Moderation analysis. Size effects should be reported and discussed.
Limitations. As a limitation, the authors should mention that they conducted a cross-sectional study, which does not allow testing whether the independent variables are risk factors for stress, i.e., whether they promote an increase in stress levels. In fact, neither pre-pandemic nor post-pandemic data were available. In addition, the authors did not ask anything about work, e.g. whether participants had the option to choose to work face-to-face or to stay at home, what kind of work the participants did, whether they had part-time or full-time jobs, etc.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The authors present and discuss their work on work-related stress during COVID-19.

The topic is within the scope of the Special Issue-Occupational Stress and Health: Psychological Burden and Burnout.

The work is systematic, complete and well organized. The results are convincing.

Sample selection could be more detailed, whether the sample of telework and face-to-face work depends on a specific occupation. Sample selection should ideally be representative.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 5 Report

The authors conducted a questionaire study on the stress level under telework and face-to-face working circumstances. The paper is well-written and the conclusions are supported by the data and analysis. This paper fits the aim and scope in IJERPH and I would recommend publication if the authors could answer the following question:

There are certain occupations that telework is not possible, and if such occupations are included in the face-to-face samples, there could be a bias towards the results as the face-to-face will have a larger set of occupations. Is it possible to filter the face-to-face samples to only the occupations that both telework and face-to-face are possible?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear authors,
I think your paper is much improved. Very good!

Please, I only ask you to report your references using the MDPI style, and to read again your paper, since few typos / minor errors are present.

Best regards

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

I find that the authors thoroughly revised their manuscript in accordance with the reviewers’ comments and suggestions. They also acknowledged the main limits of their study. All in all,  I find that they did all they could do with their data. Only two points. First, I suggest that the authors avoid putting their research within a holistic perspective, because they conducted a cross-sectional study, in which several salient variables were not taken under control and no attention was put on individual unique representations. Secondly, (p 11) unclear to me how “the model including caregiving and perceived economic threat explained the 40 % in the variance of job stress” if R squared was equal to 0.16; similarly, few lines below when caregiving and rumination accounted for 42 % of variance in stress, but R squared was 0.18.

Author Response

Please see the attachment

Author Response File: Author Response.docx