Measuring the Spatial-Temporal Heterogeneity of Helplessness Sentiment and Its Built Environment Determinants during the COVID-19 Quarantines: A Case Study in Shanghai

Round 1

Reviewer 1 Report

The COVID-19 pandemic has shown the importance of preventing anxious attitudes among the populace during epidemics. One possible method to solve this task is to make the right decisions in urban planning to ensure sustainable urban management during epidemics.

In the paper presented, the authors pose the following questions:

1. How did helplessness value on social media during the Shanghai lockdown change in a spatial and temporal dimension?

2. From the perspective of urban studies, to what extent do the built environment, housing economy, and census contribute to helplessness?

To answer these questions, the authors first examined Chinese posts with GPS locations from Weibo microblogging service in central Shanghai during the 2022 spring quarantine period and trained an NLP model to recognize nine emotions based on raw Weibo data to accurately quantify the emotions of helplessness. The built environment, housing economy, and census dimensions were then assigned to administrative streets as potentially independent variables. Finally, the authors proposed a regression model to study and analyze the correlation between helplessness and the above-mentioned independent variables.

As a result of the study, the authors were able to establish the following:

1. There was significant heterogeneity in the spatial and temporal dimensions of helplessness in Shanghai;

2. There was spatial heterogeneity with different strengths among different independent variables;

3. Different types of factors, such as housing price, construction year, population density, etc. affect helplessness to different degrees.

Thus, the authors conceptually expanded the disciplinary boundaries of traditional urban research, as well as established the correlation between subjective emotions and multi-dimensional urban indicators, aiming to effectively explore the correlation between the distribution of helplessness during the lockdown period in Shanghai and these attributes such as built environment, housing economy, and population census.

Therefore, this study provides a reference for interdisciplinary research in the context of pandemics and provides valuable insights for economists, urban planners and policymakers in the post-pandemic era.

The paper under current review may be published as presented.

Author Response

Reviewer 1: The paper under current review may be published as presented. Answer: Thank you for your careful review, kind words, and your support for our paper acceptance. Your constructive comments inspire us, and we will make more efforts in future interdisciplinary research.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper examines the impact of social emotions during the lockdown of the city of Shanghai. The concept of the paper is good, the research idea and structure are clear, and the conclusions of the study can provide decision support for urban management and planning as well as emergency response. However, the paper seems to be insufficient in terms of research methodology, discussion and summary of conclusions, and the selection of some influencing factors (independent variables) is not convincing enough, while the timeliness of some data is not good. Here are some suggestions:

 1.        The Abstract is not sufficiently summarized and further concentration is recommended;

2.        There are some language and grammatical issues that need to be revised;

3.        The research period of the paper is April-May 2022, but the census data is from 2010, which may be inaccurate to use 2010 data due to the rapid development of China's cities, which can change a lot in 12 years, especially since the population is still moving dynamically.

4.        The rationale for selecting the built environment as an influencing factor in the paper may not be sufficient. As the paper states, "Under the andatory lockdown policy in Shanghai, most citizens were isolated in the fixed and limited physical space (home)", during the lockdown period, people mainly lived in their homes and rarely had the opportunity to go out of the house, so the outdoor environment may not have a significant impact on the residents' emotions;

5.        Suggest reorganizing what should be written in Sections “5. Discussion” and “6. Conclusions” and the logical relationship between the content;

6.        Maintain consistency in references;

7.        The title of the paper does not quite match the content.

 There are some language and grammatical issues that need to be revised.

Author Response

Reviewer 2:

This paper examines the impact of social emotions during the lockdown of the city of Shanghai. The concept of the paper is good, the research idea and structure are clear, and the conclusions of the study can provide decision support for urban management and planning as well as emergency response. However, the paper seems to be insufficient in terms of research methodology, discussion and summary of conclusions, and the selection of some influencing factors (independent variables) is not convincing enough, while the timeliness of some data is not good. Here are some suggestions:

1. The Abstract is not sufficiently summarized and further concentration is recommended;

Answer:

Thanks very much, we have thoroughly summarized, revised and enriched the abstract section. Hope that the new abstract is in better shape now. Please see the Abstract section for details.

2. There are some language and grammatical issues that need to be revised;

Answer:

Thanks very much. We have revised the paper, focusing on refining the language and addressing grammatical issues.

3. The research period of the paper is April-May 2022, but the census data is from 2010, which may be inaccurate to use 2010 data due to the rapid development of China's cities, which can change a lot in 12 years, especially since the population is still moving dynamically.

Answer:

This is a crucial point, and we appreciate your reminder. Recognizing the potential inaccuracies associated with using 2010 census data in light of China's rapid urban development and dynamic population movement, we have replaced the census data with the more recent seventh census data from 2020. Furthermore, we have re-established the regression model to ensure the accuracy and relevance of our findings. Kindly refer to Section 3.2.3 for more detailed information.

4. The rationale for selecting the built environment as an influencing factor in the paper may not be sufficient. As the paper states, "Under the mandatory lockdown policy in Shanghai, most citizens were isolated in the fixed and limited physical space (home)", during the lockdown period, people mainly lived in their homes and rarely had the opportunity to go out of the house, so the outdoor environment may not have a significant impact on the residents' emotions;

Answer:

Thank you for raising this concern. We appreciate the opportunity to provide a more detailed explanation. The selection of the built environment as a key influencing factor in our paper is grounded in extensive prior research that establishes a substantial impact of built environment factors on individuals' emotions and psychological well-being during the COVID-19 pandemic (Asim et al., 2021; Cheng et al., 2021; Guo et al., 2022; Spano et al., 2021). While it is true that residents in Shanghai were predominantly confined to their homes during the lockdown, it is crucial to acknowledge that certain outings were deemed necessary, such as collective PCR testing and obtaining essential supplies.

To be more specific, since the COVID-19 infection situation has exhibited different states in various residential areas, the management and control methods under the policy were also different, not to mention the fact that the intensity of policy implementation by neighborhood committees could not be uniformly measured. In fact, except for medical staff, material sales and delivery personnel, certain residents could be allowed to walk freely on urban streets or within residential areas. For example, residents who have negative results of COVID-19 PCR tests within the entire building could walk around in the residential areas. Furthermore, in the last week of May, negative (PCR test) residents in some areas could use community passes to leave their residential areas and move freely in urban spaces or purchase goods at designated locations for a limited time. Therefore, the helplessness of residents in May is closely related to the built environment data. Addition-ally, even within the confines of their homes, external spatial factors like windows with views of greenery and the sky can influence residents' moods (Asim et al., 2021).

In essence, despite the lockdown, a vital connection persists between the external built environment and residents. This connection remains integral to understanding the emotional experiences of residents during the lockdown period. Please see the Built environment part of Qualitative description in section 4.2.2. for detail.

Moreover, we have revised the sentence “Under the mandatory lockdown policy in Shanghai, most citizens were isolated in the fixed and limited physical space (home)” to “Under the mandatory lockdown policy in Shanghai, in addition to some necessary outings, such as collective PCR testing and pickup of food supplies, most citizens were isolated in a fixed and limited physical space (home)” for accuracy in section 3.1.

5. Suggest reorganizing what should be written in Sections “5. Discussion” and “6. Conclusions” and the logical relationship between the content;

Answer:

Thank you for your suggestion, we have revised it. The discussion section presents a detailed analysis and interpretation of the study's findings, exploring the complexities and nuances of the research. It provides an in-depth examination of the exogenous variable, draws connections between variables, and offers possible explanations for effects of key independent variables. Also, the discussion presents comprehensive implications for urban studies and planning.

On the other hand, the conclusion serves as a concise summary of the entire study. It distills the key findings and insights from the discussion and reinforces their significance. The conclusion reiterates the main contributions of the research, highlights its implications, and suggests directions for future research.

In summary, the discussion section sets the stage by delving into the findings of the results, while the conclusion wraps up the discussion, emphasizing the broader implications and the overall significance of the study. Together, these sections guide the reader from the exploration of data analysis in the discussion to the summarization and synthesis of key takeaways in the conclusion. Please see 5 Discussion and 6 Conclusion section for more detail.

6. Maintain consistency in references;

Answer:

Thanks for your careful review, we have updated all citation style in text for consistency.

7. The title of the paper does not quite match the content.

Answer:

Thank you for your observation. We have made adjustments to the title. The updated title now accurately reflects the focus of our research: Measuring the Spatial-temporal Heterogeneity of Helplessness and its Influential Factors: A Case Study in 2022 Shanghai COVID-19 Outbreak

Author Response File: Author Response.docx

Reviewer 3 Report

The article presents an analysis of the dependence of the emotional state depending on the spatial location and moment in time during the 2022 lockdown in Shanghai. The hypotheses of the dependence of the state of helplessness on the factors of the built environment, housing economy and census attributes were tested.

The hypotheses formulated in the introduction were tested in detail during the study, obtaining results based on practical data. Unfortunately, the figures and tables referred to by the authors in the study could not be found in the text of the article, which greatly complicates the assessment, since a significant part of the material is missing.

There are also a number of questions to what is presented in the article.

1) On what basis were 556 words that reflect helplessness selected?

2) Why was linear regression chosen as the main model?

3) How can one claim that the model is correct if a statistically significant correlation for linear regression was found only for May? Does this mean that for each month the contribution of the independent variables can change significantly?

Author Response

Reviewer 3:

The article presents an analysis of the dependence of the emotional state depending on the spatial location and moment in time during the 2022 lockdown in Shanghai. The hypotheses of the dependence of the state of helplessness on the factors of the built environment, housing economy and census attributes were tested.

The hypotheses formulated in the introduction were tested in detail during the study, obtaining results based on practical data. Unfortunately, the figures and tables referred to by the authors in the study could not be found in the text of the article, which greatly complicates the assessment, since a significant part of the material is missing.

Answer:

Thanks for the careful review. We have included all the figures and tables in the revised version of the article. Hope they are informative and straightforward enough to meet the standard. We apologize for any confusion caused by their initial absence.

There are also a number of questions to what is presented in the article.

1. On what basis were 556 words that reflect helplessness selected?

Answer: We identified “helplessness” as the inability to defend oneself or act effectively in response to the need for daily supplies and medical treatments due to the strict lockdown restrictions. To capture the public helplessness, we manually extracted 556 keywords representing the inability to respond effectively to daily needs from the Weibo data as the “helplessness emotion lexicon.” We then annotate them into the NRC Emotion Lexicon format (Appendix Table A1). Please refer to Section 3.1 for more details.

2. Why was linear regression chosen as the main model?

Answer: The choice of linear regression as the primary model stems from its interpretability and support in existing literature. Linear models offer the advantage of providing clear insights into the influence of independent variables on dependent ones. Moreover, given the focus of this study on citizens' emotions and related factors, the decision to employ linear regression aligns with established practices in similar research. The rationale for this choice is further elucidated in the subsequent table. Please refer to Section 3.3 for details.

3. How can one claim that the model is correct if a statistically significant correlation for linear regression was found only for May? Does this mean that for each month the contribution of the independent variables can change significantly?

Answer: Thanks. The significance of the linear regression model specifically for May can be attributed to the dynamic nature of the contributing factors during different stages of the COVID-19 lockdown in Shanghai. As indicated by previous research (Hu et al., 2023), factors such as the intensity of lockdown measures, the content of restrictive policies, and the level of people's mobility varied across different urban areas and within different stages of COVID-19 transmission. These complex factors are intertwined, resulting in differences between the correlation models in different months of Shanghai's lockdown period. Specifically, due to the significance of the May model of civic helplessness (R²=0.214) (Table A5), this study currently uses the May model as a case study to explore the correlates of civic helplessness concerning the built environment, housing economy, and census dimensions. In contrast, the April model (Table A4), featuring the same combination of independent variables, is less significant than the May model, showing the different influences of the independent variables in the April and May models. This might reveal that, other than the combination of independent variables in this study, helplessness in April is largely influenced by complex and comprehensive factors at the early stage of the epidemic such as the interval of lockdown periods, the varying number of newly confirmed cases, the distribution of and the considering of human psychological fatigue, while in May, helplessness becomes more affected by stable attributes such as spatial, census and economy, contributing to the correlation analysis. Please see 4.2.2 section in Results and 5.1.1 section in Discussion for more detail. (See Table A4. The OLS model results in April. and Table A5. The OLS model results in May.)

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Thanks to the authors for their efforts to revise the paper, I have no further comments.

Please make sure the wording and grammar are correct.

Reviewer 3 Report

The article has been significantly revised. Responses to reviewers' comments are satisfactory. The new material provides a correct presentation of the methodology and interpretation of the research results.