Upholding dignity during a pandemic via Twitter

Comments to the authors by reviewer 1 (Meiko Makita)

The paper reads fairly well and offers insights into how dignity was perceived and discussed during the early days of the pandemic, particularly in relation to older adults and various societal concerns. I believe the manuscript can be further improved. Here are some revisions I suggest for strengthening the overall quality of the manuscript.

1. Form and style.

• Please be consistent with either British or American language. Throughout the manuscript you use both (e.g., ageing, aging).
• In-text citations are presented in at least two different formats.
• Please review the references list.
• On page 3, par 5: (i.e., older adults, 2SLGBTQI+, Indigenous people, immigrants, etc.) it should be ‘e.g.’ without ‘etc.’. t
• The acronym LGBTQI+ is well-known, however I do not know what 2S stand for, perhaps it would be better to spelled it out.

Response: Good points. We brought consistency to “ageing,” used Canadian English throughout, and reviewed the format of in-text citations. Instead of using evolving acronyms that vary across geographies, we refer to “sexually and gender-diverse communities.”

2. Section Theoretical Background/Social Media and Marketplace sentiment

I was surprised the authors only covered very broad social media studies and did not include any study on older adults/ageism and social media, such as the two I shared below,

Reuben Ng, Nicole Indran, Hostility Toward Baby Boomers on TikTok, The Gerontologist, Volume 62, Issue 8, October 2022, Pages 1196–1206, https://doi.org/10.1093/geront/gnac020

Elliott, R. (2022). The ‘Boomer remover’: Intergenerational discounting, the coronavirus and climate change. The Sociological Review, 70(1), 74-91. https://doi.org/10.1177/00380261211049023

Response: Thank you for the helpful suggestion. We have reviewed the studies you shared and agree they offer valuable insights into older adults and ageism in social media. We have incorporated relevant findings (Archambault et al., 2022; Bascu et al., 2022; Clancy et al., 2021; Elliott, 2022; Lothian & Philp, 2001; Ng & Indran, 2022) into the manuscript, particularly in the literature review and discussion sections.

3. Methodology

The method appears to have followed a systematic approach and the authors state that their study is reproducible, which is a positive aspect in terms of rigour. In the Discussion section the authors mention:  “Finally, by analyzing the sentiment and engagement with different topics, policymakers can understand what issues are most important to the general public and which issues may create conflict or reduce trust in government or decision-making authorities.”

However, no sentiment analysis of the tweets data set was conducted. Could you please either conduct a proper sentiment analysis or explain what is that you mean by ‘analysing the sentiment’?

Response: We removed this from the manuscript.

I suggest moving the text that is currently in the first paragraph of Results (below) to the Methods section, when ethics is introduced:
 
Social media studies often include quotes to give participants a voice, illustrate ideas to build trust, foster replicability, or lend evidence for claims. However, evolving ethical standards attend to the unintended consequences of reporting individuals’ tweets and express concern about the discoverability of messages posted voluntarily in public forums (Mason and Singh, 2022). Accordingly, we honour individuals’ privacy and do not reproduce user IDs or tweets. In contrast, we refer to companies, organizations, and institutions that use Twitter for public relations to share views with stakeholders.

Response: Thank you for this suggestion. We moved this portion of the text to the Methods section.

I’d also further elaborate on the ethical implications of using publicly available online data. There is a lot of debate on this issue and the authors should engage with this literature and further discuss their stance on using publicly available online data.
Please see for instance: Stommel W, Rijk L de. Ethical approval: none sought. How discourse analysts report ethical issues around publicly available online data. Research Ethics. 2021;17(3):275-297. doi: 10.1177/1747016120988767

Response: We included the suggested Stommel reference in this section and elaborated on the ethical implications of using publicly available online data.

The method employed is not discourse analysis but thematic analysis. This is also supported by the name of the subsection ‘thematic analysis and word clusters’. Perhaps remove from the keywords ‘discourse analysis’ and only retain ‘discourse’.

Response: We corrected this in the text by removing discourse analysis.

4. Results

Tweets with most engagement – is this about number of likes and retweets? Could we have more data on this for each of the top-10 accounts?
Also, since the audience would be not only from North America, probably best to specify this is USA’s Democratic Party.

Response: We have included engagement details and made our descriptions more precise.

In comparison, with the other four themes, ‘recognise dignity as a fundamental right’ is not as developed. Or is the first theme the overarching theme and the other four subthemes?

Response: The first theme is general, often providing context to introduce more specific themes. We added a transition sentence to improve clarity.

Consider changing the themes to an ‘ing’ verb form, example: Preserving the dignity of at-risk populations.

Response: Good idea – thanks!

The section ‘Tweets focused on older adults’ seems to be a little underdeveloped, especially since is a focus of the study. In here, the authors could perhaps introduce examples of what was said in tweets, as they did with the other themes.

Response: We thank the reviewer for this comment and wish to clarify that the study focused on dignity. A secondary aim of the study was to examine a subset of posts that focused on older adults and dignity. We added text and research questions to the introduction and abstract to clarify this.

5. Discussion

Once the authors expand on their theoretical background section, they would be able to also strengthen their discussion.

The concept of ‘healthy ageing’ is only introduced at the end of the manuscript – if this is important try to bring it up much earlier on (intro, background).

Response: In the Introduction, we discussed how older adults were discussed and how this may infringe on dignity: “Indeed, in the context of COVID-19, initially, the discourse surrounding older adults living in long-term care homes focused heavily on the loss of dignity in healthcare and in dying (Carrieri et al., 2020; Colombo, 2021; Kowsalya & Sundara, 2021, Vellani et al., 2021). In general, much of the discourse was ageist, promoting notions that all older adults need protection, are vulnerable and should be socially isolated (Fraser et al., 2020, Lagacé et al., 2020) –all factors which infringe on human rights and dignity.” This is important at the main focus of the paper is DIGNITY and secondary focus dignity and aging.
We mention healthy aging in the last paragraph of the discussion to emphasize how changing the discourse around older adults to a more positive orientation while diminishing or removing ageist perspectives will support healthy aging in this population.

Consider adding a ‘conclusion’ section.

Response: Thank you for this suggestion, we added a conclusion section: “This study analyzed Twitter posts about dignity during the first year of the COVID-19 pandemic, with a focus on the intersection of dignity and older adults. Using text analytics and interpretive methods on a cross-national sample, we identified five major themes: recognizing dignity as a fundamental right, upholding the dignity of essential workers, preserving the dignity of at-risk populations, preventing cascading disasters, and attending to death dignity.
Our findings emphasize the widespread discussion of dignity-related challenges faced by essential workers and populations experiencing vulnerability, particularly older adults. This underscores the urgent need for policies that address these challenges to promote human dignity. Social media has captured real-time public sentiment and provided insights that inform policy and practice. In summary, this research improves our understanding of public perceptions of dignity during a global crisis, emphasizing the importance of policies that uphold dignity, especially for those most affected by the pandemic.”


Comments to the authors by reviewer 2 (Hannah Loret)

Title and Abstract:

• The summary of the background and purpose of the research should better articulate the research question and the need for this study to have been undertaken.
   
• Sample size is clear, but how the data was identified and sourced could be covered here.

Response: Thank you for this suggestion. We edited the background and purpose statement to clarify the need for the study and the questions we set out to answer.

• The conclusion is very descriptive. What is the main finding (it is clear however why we should lament the unacceptable situation for older adults)? This could show a stronger critical engagement with the results of the study.

Response: To refine the conclusion in response to this comment and that of the other reviewer, our conclusion now reads: “Conclusions: Within the discourse, dignity emerged as a right and a policy target (improving or maximizing dignity in various groups). Further, the posts focusing on older adults and dignity demonstrated that the impacts of COVID-19 disproportionately impacted older adults, particularly those with greater care needs and lower socioeconomic status.”

Introduction:

• Referring to structurally underserved ‘vulnerable’ people as ‘creatures’ (p. 3 para 5) could be better expressed here, by choosing a different quotation to make the point. As it stands, the way it is currently expressed risks repeating the terminology used to marginalise the groups mentioned. 

Response: Thank you for this comment. We rephrased the sentence: “In extreme cases, dignity violations may undermine personhood, as individuals risk being treated as “not human or less than human–as a thing or instrument or subhuman creature” ( Kateb, 2014).”

• Later in this same paragraph, there is the claim that to consider older people as all vulnerable is ageist, but earlier in the paragraph the authors point to older people as a vulnerable group. Perhaps this could be mitigated by suggesting older people, ‘immigrants etc.’ are potentially vulnerable, allowing for slightly more nuanced use of the term?

 Response: Thank you for noticing the discrepancy. We rephrased accordingly: “For these reasons, marginalized individuals or groups may be at greater risk of experiencing vulnerability ( i.e., older adults, children, immigrants, Indigenous people, low-income individuals, and sexually and gender-diverse communities) and are at a greater risk of losing their dignity”

• P. 3 paragraph 6 I am unsure why Latour is being brought in. Are his methods being used? This reference to his work or approach doesn’t seem to be upheld in this section or signposting to the methodology, beyond saying that research into media discourse and public opinion (which are two different things – perhaps a distinction could be made?) has a similar investment in making meaning.

 Response: We rephrased this section to clarify why Latour is used and to distinguish between media discourse and public opinion: “Sociologist Bruno Latour (2007) explores how language and representation shape social interactions and worldviews. His perspective is relevant to our study of social media discourse on dignity during the COVID-19 pandemic, as we analyzed how users interact through tweets, using symbols and narratives to construct and express their views on dignity. Similarly, researchers who examine media discourse and public opinion also focus on the symbols and reasoning devices used to present and discuss issues, such as metaphors, exemplars, catchphrases, depictions, and visual images, as well as the roots, consequences, and appeals to principle that suggest how to think about an issue ( Gamson & Modigliani, 1989).”

• (P.4) I would like to see more references and conclusions drawn in this field which justify the need for the study from sources which aren’t written by this article’s authors.

 Response: Several references have been added to the introduction.

• I understand on a basic level why this approach has been chosen, but I understand less what the current literature in this specific field looks like and what contribution to the field this article makes. It could be much more explicit in the theoretical background exactly why this study was needed from both a theoretical and empirical perspective.

Response: We added details and references to the introduction, just before text which clarifies our research questions: “Scholars argue that dignity is a fundamental aspect of human rights, crucial for realizing other rights. Theoretical work emphasizes the intersectionality of dignity, highlighting how it is affected by various factors such as age, gender, socioeconomic status, and cultural background. Marginalized groups, such as older adults, are especially at risk of having their dignity violated ( Archambault et al., 2022; Bascu et al., 2022; Clancy et al., 2021; Elliott, 2022; Lothian & Philp, 2001; Ng & Indran, 2022).”

• I’m unclear from the introduction what the exact aims or hypotheses of this study/article are. Can these be stated explicitly?

Response: We added our research questions in the Introduction: 
“Therefore, the following research questions guided our investigation:
Research Question 1: What issues of human rights and ethical considerations are highlighted in high-engagement tweets on dignity?
Research Question 2: What themes emerge from dignity-related tweets that reflect broader societal concerns about health, social justice, and public policy during the pandemic?
Research Question 3: How do dignity-related tweets address the challenges older adults faced during the COVID-19 pandemic?”

• Why were older adults chosen over any other demographic group for the focus here? Please expose the gap in the literature and the rationale for this choice.

Response: In response to comments from both reviewers, we added details on our rationale for focusing on older adults in the introduction: “The COVID-19 pandemic brought increased attention to the concept of healthy ageing, which involves maintenance of the ability to function well as people age, while preserving dignity. Ensuring older adults can live with dignity and respect is crucial to promoting their overall health and well-being (WHO, 2015; Beard et al., 2016; Sixsmith & Sixsmith, 2008). The WHO's "Decade of Healthy Ageing" initiative (2021-2030) emphasizes the need for comprehensive care and long-term support systems to improve the lives of older adults during and after the pandemic (WHO, 2020). Additionally, a life-course model for healthier aging, proposed by Jowell, Carstensen, & Barry (2020) in The Lancet, highlights lessons learned during the COVID-19 pandemic and emphasizes the importance of supportive policies to support the health and dignity of older adults. The main goal of this study was to examine how the concept of dignity was discussed on Twitter (X) during the COVID-19 pandemic. Secondly, given that older adults are at a greater risk of having their dignity violated (Clancy et al., 2021; Lothian & Philp, 2001), this study targets the intersection of dignity and older adults in Twitter posts.”

Methods:

• (Para 2, p. 4) I disagree that Twitter is a highly inclusive platform. What about people who have literacy issues or lack digital skills, internet access, or other unmet access needs that prevent them from engaging with it?

 Response: We agree with the reviewer and rephrased this paragraph to state the benefits and pitfalls of using the Twitter  (X) platform: “Our study examined original messages from Twitter users, a one-to-many platform where people share information in a relatively synchronous or condensed time frame ( Lamberton & Humphreys, 2022). Twitter data allows researchers to gain insight into public opinion, including its volume, sentiment, and expressions. Despite Twitter's efforts to be inclusive, it is not accessible to everyone due to literacy issues, digital skill gaps, and limited internet access. Although the platform has a diverse user base, its primary users generally fit the WEIRD criteria (Western, Educated, Industrialized, Rich, and Democratic) ( Septiandri, Constantinides, & Quercia, 2024).”

• (Para 3, p. 4) why were English-language tweets chosen over another language? The rationale should be clearer for this.

Response: To clarify why English-language tweets were chosen, we added, “Our cross-national sample of English tweets was curated with a focus on researcher expertise, bolstering the accuracy of our interpretations. However, we recognize limitations in extending these findings to other languages.”

• (P. 4) can you speak a little more about how this data was analysed, to signpost to the results section? The methods used are clear and seem appropriate, but I’m struggling to see how they fit within the overall aims of the project.

Response:  The data were analyzed thematically to show how comments on dignity and older age circulate on public platforms and reflect broader societal issues related to the pandemic, dignity and age. Our multi-method approach blended text analysis algorithms and human interpretation – leveraging the strengths of the research team. The methods section was revised to eliminate jargon and improve clarity.

• (p. 4) could a rationale be given for why this method was chosen over another, and how any bias may function within it?

 Response: In the discussion, we addressed this comment in our limitations paragraph: “The rise of social media has opened new avenues to explore public opinion, yet it has limitations. First, social media users do not represent the general population's views due to self-selection and non-random participation. Therefore, combining this method with other research methods may be necessary to fully understand public views on human dignity. Second, while individuals and organizations control the messages they create, the algorithms and platforms used significantly influence how their intended audience sees and receives those messages ( Kozinets & Gretzel, 2021). Third, information shared on social media can be contaminated with misinformation and tainted by stereotypes, so it is essential to be vigilant in promoting accurate and credible information and be aware of potential biases.”

• I find the design coherent, but the research questions need to be more clearly stated earlier in the article so that perhaps the methodology section conveys more clearly how these will be answered.

Response: We included research questions in the Introduction to better guide the reader. 

• Some of the information in the ‘results’ section would be better placed in the ‘methodology section’, as it is describing the methodology rather than the results. I’m particularly thinking of the last section of the final paragraph on page 5 where this is a description of methodology rather than results. Methodological explanations at the beginning of the results section could be incorporated into the methodology, allowing more room for the thematic structuring of the results sections.

Response: Thank you for this suggestion. We moved this portion of the text to the Methods section.

Results:

• A short summary of the main results which relates to the research questions (as well as methodological focus) would be helpful.

Response: We thank you for this suggestion and added the following summary at the beginning of the results section: “The findings from this study are structured around five major themes identified through an analysis of 1,946 original tweets about dignity during the COVID-19 pandemic. These themes include recognizing dignity as a fundamental right, upholding the dignity of essential workers, preserving the dignity of at-risk populations, preventing cascading disasters, and attending to death dignity. Each theme is explored in detail below, highlighting the unique challenges and discourses associated with each aspect of dignity.”

• I particularly like the final paragraph of page 5. The tone is pleasantly authoritative.

 Response: Thank you.

• (P.8, final paragraph) I’m unsure why migrants in South Africa have been chosen to demonstrate this point. Could this point be made clearer with reference to the points in the preceding sentence, for example by explicitly explaining how this relates to questions of race/wealth/housing conditions?

 Response: We agree that additional context is needed in this section. We have provided this in our revised version: “These inequalities came to the fore with discussions about unequal access to vaccines. For example, one tweet advocated for universal access to vaccines for all citizens of South Africa, including migrants. Similarly, hashtags demonstrated solidarity with the social equity priorities of groups like Black Lives Matter, Latinos, Pride, and Veterans. Next, we discuss the overarching principle of recognizing dignity as a fundamental right, echoed in other themes focusing on specific populations.”

• (p.9, para 4) I’m unclear how the study data demonstrates that by ‘taking these steps, the public can support and empower essential workers and help them continue providing vital services to society’. My understanding is that the data demonstrates a public acknowledgement and will to improve the working conditions of essential workers but the steps to this are unclear, and it is unclear how a Twitter discourse actively empowers workers or helps them to deliver services. Perhaps a closer reading of the data gathered could make this clearer, in terms of how the public demonstrated support for the dignity of essential workers through their tweets, then linking this to any related targeted activism which followed for those particular groups of working people?

 Response: We revised this section to state the results and potential impacts for healthcare workers clearly. We also added a reference to support the view that positive/supportive posts of essential workers have a positive impact on their engagement and well-being (see Shan et al., 2022): “Finally, other messages championed the dignity of frontline and healthcare workers by calling for respect for their rights and dignity as individuals—and avoiding discrimination or harassment. This type of public support and can empower essential workers and motivate them to continue providing vital services to society (Shan et al., 2022).”

• (p. 9, para 5) This section could also benefit from some closer reference to the empirical data from the study to strengthen its statements and enrich the points made. I also questioned the use of the term ‘at risk’ when the term vulnerable had been used before. Perhaps it would be good to stick to one of these terms or to give a reason for the change in terminology?

Response: We bolstered the section with examples and revised the text to use “at risk” throughout. 

• (P. 9, para 6) Is there a reason that this section chooses to focus more closely on older adults as a structurally underserved population? This needs to be more explicit.

Response: Thank you for this comment. We revised the introduction to state upfront the importance of this specific population as underserved and the focus on the intersection of older adults and dignity during the pandemic. We’ve added references to set up our research questions and elaborate throughout the paper.

• (P.9, para 7) The first sentence here makes it sound like older adults and women are two completely distinct groups, and that older women are not included in older adults. Please reword to clarify that older adults includes women, though (younger?) women experiencing domestic abuse, precarious housing, pregnancy-related issues etc. may be affected by unique gender-related challenges.

Response: We revised this paragraph to address this comment: “Older and younger women faced unique challenges that threatened their dignity as access to necessary healthcare services exacerbated existing issues. Many older women took on additional caregiving roles during the pandemic without adequate support, adding to the burden of unpaid domestic work and putting stress on their physical and mental well-being (Age International, 2021). Additional challenges included pregnancy, parenting responsibilities, domestic abuse, or homelessness, and addressing these issues may require special attention and support. For example, the United Nations Population Fund (UNFPA) Kenya announced a collaboration to provide “dignity kits” to girls living in informal settlements.”

• (P.9, para 8) The paragraphs here are very short and therefore lose a sense of coherence. They could be joined without losing the sense of the argument, and this would also help to strengthen the use of empirical data to evidence the points being made (I would again like to see more information from the empirical data to back up the statements being made here).

Response: This section expands several paragraphs to provide additional context to the statements. We summarize the essence of tweets to reduce privacy and traceability concerns related to verbatim reporting.

• (P.9, para 8) Needs a reference or several references for ‘critics describe…’

Response: We added two references for this statement: “Critics describe how well-intentioned COVID mitigation strategies suffer from blind spots of missing information about unresolved social problems, neglecting risks, compounding the harm, and leaving dignity in peril (Dorn et al., 2021; Kharas & Dooley, 2021).”

• (p. 10, para 1) Readers who are not in the USA may not be familiar with the political parties or political structure there, or the exact details of the Capitol attack in January 2021. If this is to be included, it will need to be qualified by some more information about the USA political system which links to the point being made about propaganda/misinformation and discriminatory policies.

Response: We revised this section and provided examples in brackets to clarify for any reader: “Dysfunctional politics can erode individual dignity by endorsing discriminatory policies, such as voter suppression laws and travel bans targeting predominantly Muslim countries. Additionally, propaganda campaigns, which promote biased narratives for political gain, and misinformation, such as false claims about voter fraud or the COVID-19 pandemic, undermine public trust and exacerbate division. On social media, sharp criticisms and personal attacks between American political figures often provoke equally harsh responses, creating an environment where public discourse lacks respect. Dignity emerged as a central theme in the polarized political climate leading up to the U.S. November 2020 presidential election. Thus, advocating for inclusive and respectful politics that uphold the dignity and rights of all individuals remains crucial.”

• On this point, I think discriminatory policies and propaganda/misinformation are two separate points. Perhaps these could be teased out briefly with reference to empirical data to justify these statements.

Response: Thank you for this excellent suggestion. Our revised paragraph above shows that these points are distinct.

• (P. 10, para 4) It is important to acknowledge that although many healthcare systems operate both publicly and privately, and have social security, healthcare system design varies and the example given from Texas will be unique to the case in the USA which should be clearer – for example in Scotland, ‘insurance status’ would mean something very different in terms of healthcare, and most of the population is able to access primary and secondary care free of charge.  

Response:  We agree with the reviewer and have added two sentences to clarify that differences will depend on context: “Healthcare system designs differ across countries and regional jurisdictions, influencing access to primary and secondary healthcare services without financial burden. These variations can significantly affect individuals' ability to uphold their dignity.”

• (p. 10, para 6) it is unclear why we are suddenly speaking about end-of-life care. This could be better linked to the empirical data, and could be linked to the above paragraph with a signposting sentence to improve the flow of the article.
  (p. 10, para 6) It is unclear why we are looking at Kenya and India for these mentions, the connection could be more explicit and better linked back to the concerns of the overall analysis (if the article wants to concentrate on meaning, there will need to be more critical appraisal or expansion on possible meanings here as elsewhere).

Response: We revised the text on page 10, paragraph 6, to address these two comments from reviewer 2. The revised text includes an improved explanation for the points raised and several references for the details around dignity, death, and dying: “Respecting an individual's dignity during their end-of-life can be crucial to providing decent end-of-life care (Kennedy, 2016). Dignity is a quality or state of being worthy of respect and honour. As individuals approach the end of life, they often wish for a “good death” that is peaceful, pain-free, and by the individual's wishes and values (Meier et al., 2016). Respecting an individual’s dignity during end-of-life care is crucial for a peaceful transition (Meier et al., 2016). Yet, several barriers emerged during the COVID-19 pandemic, including reduced access to loved ones, overwhelmed healthcare systems, lack of personal care, the need for personal protective equipment, and distressed healthcare workers (Hernandez-Fernandez & Meneses-Falcon, 2021).
Additionally, a Kenyan argued that providing protective suits might allow people to give their final respects to their loved ones with dignity. Finally, some followers of an Indian spiritual leader tweeted claims that they had received a divine cure for the deadly virus, telling of healing without medicine. The COVID-19 pandemic highlighted the need for adaptive practices that honour cultural rituals while safeguarding public health, revealing the deep connection between culture, spirituality, and the dignified handling of death.

• It is not until p. 10 we are told why older adults are a particular focus. This rationale is strong and needs to figure much earlier in the article.

Response: We added details to the introduction to clarify our choice of older adults as a focus.

Figures and Tables:

• I find that figures and tables are clear, easily interpreted, and well-formatted.

Response: Thank you.

Discussion:

• The start to this section is strong, and the tone is appropriate. The summary of the main themes is particularly helpful.

Response: Thank you.

• There needs to be some discussion in this section of existence of opposing views to the ones presented. The wider context around the findings and any other research which may contradict these findings need to be explored, and linked back to the research gap that this study was filling.

Response: Thank you for this suggestion. Our study was not designed to directly challenge or contradict existing research findings. Instead, our contribution lies in giving voice to a largely overlooked or unaddressed area—citizen talk of dignity on Twitter (social media). The revised introduction and theoretical background sections clarify how our work aligns (or does not align) with others’ work.

• The limitations are clear, perhaps the strengths need to be better situated within the available literature and existing knowledge on this topic.

Response: Thank you for this suggestion. We’ve added references to bolster our limitations section.

• Future directions for the research and practice are clear and meaningful.

 Response: Thank you.

References:

• Some inconsistencies for referencing (particularly of The Lancet - Lancet, T. as author of one editorial, please correct this in line with style guidelines). Please ensure consistency and revise these.

Response: Thank you. We revised the inconsistent referencing.