Public Interest, Risk, Trust, and Personal Protective Equipment Purchase and Usage: Face Masks Amid the COVID-19 Pandemic

Round 1

Reviewer 1 Report

The Survey and approach for qualifying respondent data appears to be reasonable. Incomplete surveys were excluded, two pilot surveys were used to improve questionnaire design, 1054 surveys representing 27 mainland China provinces were qualified for use in data evaluations. Dependent variable were mask purchases and uses.

Independent variables were 1) public interest based on keywords used in internet searches, 2) risk aversion of respondent based on an experiment of reward and gamble to assess a respondents' general aversion to risk, and 3) trust, using a Likert scale. A variety of COVID associated variables and demographics was also collected for data analysis.  Data was related with two COVID periods – Period 1 outbreak and Period 2 recovery.

 

 

Suggested minor editorial changes throughout. For example:

line 62 change "evolvement" to "evolution"

line 76 change "experiment" to "experimental"

line 88 change "2020." to "2020?"

line 94 change "data on public" to "data from a public"

other minor grammatical changes needed throughout

 

 

 

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript is well and clearly written, and I believe it should be accepted for publication without changes.

Manuscript deals with the face mask demand during and after the COVID-19 surge in China. The survey focused on mask purchase and mask usage in the first and second quarters of 2020.

Attitudes towards wearing face masks during the COVID-19 pandemic were highly variable and most often posed a controversial issue, as did vaccination, with pronounced regional variations. China is a populous country, and the authors considered regional specificities. Understanding these views is of primary public health importance, taking into account the duration of the pandemic and the consequences it leaves around the world.

The author's approach was original and compelling: a preliminary screening of Internet searches on COVID-19 showed the specific public interest in the research topic in particular cities in China. Risk aversion and trust were then taken into account, plus COVID-related- and other demographic variables. The exceptional quality of the manuscript is the original economic analysis. The study covered the 'outbreak' and 'recovery' periods (Jan to Mar and Apr to Jun 2020, respectively).

The study is complete, well documented, and clearly presented. It is a significant step forward in public health aspects of the current pandemic. The discussion is comprehensive and adequately compared to other similar studies. I do not think that changes or additions are needed, but it would be interesting to follow up on the same issue, i.e., update for 2021.

The conclusions are consistent with the results and objectives.

The references are appropriate.

I don't have any additional comments on the tables and figures.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Please improve introduction in the international context

Could you also present better the results in a discussion 

Very appreciate semplification in results presentation (the scientific contribution is average, it would be an asset to simplify some sentences to reach different stakeholders) 

Moderate English changes are required.

The paper is interesting, its purpose and design are quite clear and the results and conclusions presented could be have a relevat interest for reader particularly for people who operated in policy. In fact the study considers public interest in COVID-19 related issues as well as individuals’ risk perception and trust in society in their demand for face masks during the pandemic. The research is useful for policymakers to assess the creation and termination of temporary legislation to help manage the value chain of PPE during a major public health crisis. The scientific contribution is avarage. The paper needs a moderate english review. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The authors analyzed the purchase and use of personal protective masks during the COVID-19 pandemic (health crisis) and during convalescence only in China. In this system, it is difficult to conclude the possible translation of the results to other countries, it would be good to plan multi-center studies. The author emphasizes that China is a large country and the study was conducted on only 1054 respondents - which is the control group that should be reported. I suggest the authors respond to the comments.

-The study assessed the factors influencing the pyrchase and use ( wear) of masks in the period of the risk of the COVID-19 pandemic

- the influence of the measurement was researched: risk- aversion rate concerning the demand for face masks, the role of the public risk management factor was also determined

-the public confidence in the risk management system was examined during and after the epidemic

The objectives of the study were structured correctly, although not very clearly.

The research is not innovative, it is known that the period of the pandemic generates the need to protect against COVID-19 disease ( wearing masks during the period was an obligation), while in the post- pandemic period, the society gradually resigns from wearing them.

In the methodology, I would suggest the use of detailed questionnaires

The conclusions correspond to the purpose of the work

The references used and cited correctly

Tables and figures are correctly constructed

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 5 Report

In this paper the authors conduct a large scale survey relying on memory questions to investigate the purchase and use behaviour of face masks during the first six months of the COVID-19 pandemic in China. In formulating their survey questions, the authors separate the first six months after the outbreak into two parts and ask participants how many masks they purchased and wore in either period. They combine the survey answers to these questions with other survey items, i.a. the price of masks, public attention to the pandemic, individual trust, etc.

Most interestingly, the authors find no price sensitivity in mask demand and usage in the first part of the pandemic outbreak, yet a strongly significant negative effect of price changes on both of these factors in the second part. The authors then discuss other factors influencing participants’ (recollection of) mask purchase and usage.

Overall, I think the biggest contribution of the paper is the disaggregation of mask purchase and usage, as oftentimes studies focus only on either aspect. The paper can be of interest to some readers albeit some reservations about the study design, which I hope could be addressed more proactively in the conclusion section.

My biggest concern is about the timing of the survey. As the survey is collected in later summer of 2020, all survey questions concerning the pandemic outbreak period rely on the participants’ memory from that period. Surveying from respondents' memory is usually noisy because memories may be blurred and answers approximated. I understand that there is not a lot that can be done about this feature, I would hope for the authors to at least acknowledge this limitation and discuss what effect they think this may have on theits findings. For example, how reliable is a person's recollection of how often she wore a mask half a year ago?

Specific comments, ordered as I come across in the paper, not in importance.

1. Page 2, lines 82-83: Regarding also Figures 1-3, there appear to be some provinces with little response. Could you provide some more information on the regional breakdown of your data? Something like respondents per province. Relatedly, what is your overall response rate? Under proportional sampling scheme I assume I have to imagine a proportional stratified sample. If so, is there substantial variation in the response rate between the strata?
2. Page 3, Subsection 2.2: Please be explicit about how you collected the variable ``Price''. If I understand your study correctly, you used the same method here as for the purchasing & usage data, right? This would mean something like ``What did you pay for a mask six months ago?’’
3. Page 3, Subsection 2.2, paragraph (b): The risk aversion task was not incentivised, I assume? All sources you cite concerning your risk aversion scale concern incentivised risk aversion measures. Is there evidence that unincentivised risk aversion tests in surveys produce reliable results?
4. Page 4, Table 1: Concerning the variable ``Trust’’ and looking at the explanation provided under *, it appears that the highest trust level is at 1, lowest trust level at 7. In this sense you measure dis I would consider inverse the coding of the variable to actually measure trust, or to rename into distrust. This issue permeates into the results and conclusion. If in fact you measured distrust, the interpretation of your results would need to be inversed. Concerning the variable ``Health’’ and looking at the explanation provided under ***, also this appears to be inversely coded. As it currently stands, it appears to measure poor health. This issue equally carries significance into, i.e. Subsection 4.1 and others when interpreting your regression results.
5. Page 6, line 151: I was a bit confused about the formulation ``Putting purchase … periods together,’’. I understand you simply divide usage by purchase to get the share of used masks as a percentage of totally purchased masks. Please be explicit about this to avoid confusion.
6. Page 9, Paragraph (1): What I wonder about here is your discussion on the governmental price regulations you talked about when arguing that there would be no endogeneity problems about the price. As the price for masks seems to be regulated, how can it be that people pay more than CNY12?
7. Page 10, line 371: ``They might…healthier individuals.’’ Unfortunate that this was not an aspect of the survey. This would allow to check your hypothesis / explanation of the unintuitive age effect.
8. Page 14, Paragraph (1): Maybe worthwhile illustrating the convexity of the demand function graphically? This could add to the intuition of the conditional marginal effects tables.
9. Page 15, lines 432-439: How do you explain the positive squared term? This makes the demand convex, attenuating the negative effect for large prices. It appears as if individuals become less price sensitive for higher mask prices. Is this a common feature among other normal goods too?
10. Page 16, lines 501-502: Regarding Tables 2 and 3, how can this statement be made concerning public interest? If anything, the effect appears to be more robust for mask consumption than for purchase.
11. Page 17, Conclusion: You talk about ``unnecessary hoarding’’, yet I miss evidence for the presence of this in your sample. There is some more purchase than consumption, but not dramatically so. Importantly, you do not provide direct evidence of supply issues caused by hoarding, do you?
12. Page 17, lines 517-519: Your marginal effects in Tables 5 and 6 show that participants are sensitive to comparably high prices. In that sense, the price mechanism could function to prevent excessive hoarding. In fact, artificially keeping the price low could have been the catalyst to hoarding behaviour.
13. Page 17, lines 531-534: This policy recommendation has a dangerous undertone in connotation with your results. For example, given the positive effects of ``Public interest'', following this last policy recommendation may entail suppressing relevant information to the public in order to reduce mask demand. In a similar vein, reducing a person's social network would also reduce her (excess) demand for masks. Again, would this really be desirable as role for the public sector, to nudge the amount of social network a person has towards a lower number?

Editorial comments, ordered as I come across in the paper, not in importance.

1. Abstract, line 14: ``examine demand over both the’’. I would recommend reformulating into ``examine demand during both the…’’.
2. Abstract, line 15: ``and usage, allowing us’’. I would recommend breaking the sentence here, minor reformulations to make the sentence grammar correct after the sentence break abound.
3. Generally, but in particular Figures 1-3: With figures, please place the caption underneath. With Tables, please place the caption above.
4. Figures 1-3: Please separate the words in the title, i.e. Mask Purchase and Mask Usage. For Figure 3, please place the caption together with the figure. Please avoid a page break between the caption and the figure.
5. Subsection 2.3: The headline for (1) was not in bold font. Please harmonise both. Also, please capitalise the first word in a headline. I.e., for (2): ``Prices...''
6. Page 6, line 167: Maybe add the dollar price for the at-the-time exchange rate here, because it is the first time a CNY amount appears prominently in the text. You provide an exchange rate in Footnote 6, but please put this earlier than Footnote 6. Either in addition to this footnote or instead of (preference for adding another statement on the exchange rate).
7. Page 6, line 178: ``indicating risk-averse’’ Please reformulate. This sounds a bit odd.
8. Page 7, lines 198-205: This paragraph only repeats the previous paragraph. Please delete.
9. Page 7, Footnote 5: The link does not work for me, unless I add an ``l'' at the end, i.e. ``html''. Please double check and correct link provided.
10. Page 8, Equation (1): the “if”-part is a text statement, now it appears as a formula though. Please revise. Usually, this has to be placed inside a text window within your formula.
11. Page 8, various instances: Please consistently refer to variable t as either time or period.
12. Page 8, line 272: Please do not put a headline at the bottom of the page. Also, I would expect something like ``Results’’ as title of this section, not ``Discussion’’. Maybe Results and Discussion could be an option.
13. Page 10, line 351: ``usage at the 1% and 10% levels,’’. In the table the latter appears as 5%. Please check and correct.
14. Page 10, line 359: I assume the 9.8 refers to Table 3. Then, however, I think it should be 9.5, seeing the table.
15. Page 11, Table 2: Can Table 2 appear closer to when it is first mentioned? Also, is it possible to not break the table across pages? This is true for other tables as well.
16. Page 15, line 446: Please avoid the repetition of the word ``role’’ in the first sentence. Maybe just ``Risk aversion played a similar role as in Period 2.’’
17. Page 16, line 460: Instead of ``post-pandemic period’’, maybe rather use ``post-outbreak period’’. The pandemic status of COVID-19 has not seized after the first quarter of 2020.
18. Page 16, line 483: In the table I see different values than 5.4 and 3.9. Are these maybe rounding errors? Please check.

Author Response

Please see the attachment.

Author Response File: Author Response.docx