Upholding dignity during a pandemic via Twitter

Background: This article investigates how people invoked the concept of dignity on Twitter during the first year of the COVID-19 pandemic, with a secondary focus on mentions of dignity in the context of older adults and ageing. Methods: We report the results of a study that combines text analytic and interpretive methods to analyze word clusters and dignity-based themes in a cross-national sample of 1,946 original messages posted in 2020. Results: The study finds that dignity discourse on Twitter advances five major themes: (a) recognize dignity as a fundamental right, (b) uphold the dignity of essential workers, (c) preserve the dignity of at-risk populations, (d) prevent cascading disasters that exacerbate dignity's decline, and (e) attend to death, dignity, and the sanctity of life. Conclusions: Moreover, messages focusing on older adults lamented the disproportionate death toll, the terrible circumstances in long-term care homes, the added impact of suspended meal delivery services and the status of older people living below the poverty line.


Introduction
Dignity is a concept considered by many to be a fundamental human right, reflected in many legal and ethical frameworks around the world, including the Universal Declaration of Human Rights, which states that "all human beings are born free and equal in dignity and rights" (United Nations, 1948).Those who uphold dignity as a human right often argue that it is an essential component of a just and fair society and should be protected and upheld by governments and other institutions.
This paper investigates how people on social media perceive and express their views on human dignity in the context of the COVID-19 pandemic.To do this, we analyzed a large sample of Twitter posts and comments from the first year of the pandemic to understand how people respond to situations that may involve the violation of dignity.Many studies have shared insight into dignity, drawing perspectives from fields including law (Cunningham, 2020), public policy and politics (Nilsson et al., 2022), bioethics (Faneye, 2014), and healthcare (Fernández-Sola et al., 2012).However, studies usually focus on expert views, not how the general public expresses thoughts about dignity.
Examining data from social media has many advantages, as it provides unprompted real-time snapshots of how people think and feel about a topic (Belk, Fisher, and Kozinets, 2013;Berger et al., 2020).It also provides a large and diverse user base from different backgrounds, ages, and locations.Moreover, access to a wide range of archived data (including text, images, and videos) can deliver a large sample of data that is: timely, convenient, cost-effective-and, most importantly -rich and nuanced insights into perceptions.Organizations can then use this information to formulate policy, improve their products and services, and develop more effective marketing and communication strategies.
In the following sections, we describe the theoretical foundations of our approach and report the results of a study of how people talked about dignity on Twitter during the pandemic.Finally, we discuss the value of studying public perceptions of dignity and how crowdsourced views on dignity may inform policies and practices that aim to promote and protect dignity in various settings, such as workplaces, healthcare facilities, and other social settings.

Theoretical background Dignity
While it is clear that dignity and human rights are closely associated, it is important to consider that dignity encompasses: how a person feels (their self-worth) and the context in which they live (i.e., psychological, social and cultural factors) (Mégret & Hoffmann, 2009).Furthermore, dignity is socially constructed as it develops and is dependent on social interaction between individuals or groups, and it is also determined by and intersects with other human rights (Mégret & Hoffmann, 2009).For these reasons, it is likely that individuals or groups that are already marginalized or considered 'vulnerable' (i.e., older adults, 2SLGBTQI+, Indigenous people, immigrants, etc.) are at a greater risk of losing their dignity and to be treated as "not human or less than human-as a thing or instrument or subhuman creature" (Kateb, 2014).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.The impact of COVID-19 on human dignity has been negative, but it has also raised awareness about dignity and the importance of including it as a priority in health system reform (Galea, 2021).

Social media and marketplace sentiment
Sociologist Bruno Latour (2007) examines how people and groups within a society make sense of the world and communicate their understanding to others.He studies how language and other forms of representation shape social interactions and relationships, and how they influence the way people construct and communicate their worldviews.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 and Modigliani, 1989).
With increasing convergence of online and offline realms, digital methods of studying communication behaviour have become helpful in gaining insight into people's experiences (Caliandro, 2018) and sentiments or shared emotional dispositions (Gopaldas, 2014).As a result, researchers have increasingly turned to social media conversations to understand what people discuss and express.Users post messages on social media that offer a "window into experience" as they share their expertise, exchange questions or concerns, and reinforce each other's positions (Eriksson and Salzmann-Erikson, 2013).This approach has the advantage of "listening in" on conversations and avoiding certain response biases that may occur when asking questions in interviews or surveys (Rappaport, 2011).
Applications of social listening in public health and well-being contexts include studies of patients' reactions to knee replacement surgery (Pitt, Mulvey, and Kietzmann, 2018), gamblers' reflections on problem gambling (Brown, Caruna, Mulvey, and Pitt, 2021), expressions of stigma against people with dementia (Bacsu et al., 2022), public attitudes towards vaccines (Fazel et al., 2021), job satisfaction and turnover (Lam, Mulvey, and Robson, 2022), and retirement travel planning (Mulvey, Padgett, and Lever, 2022), using diverse online data sources including discussion forums, review platforms, Facebook, Twitter, and Reddit.In addition, social media methods are well-suited to collecting and analyzing data rapidly in times of crisis or responding to emerging trends, such as the COVID-19 pandemic (Picone et al., 2020;Reid and Duffy, 2018).Furthermore, advice sharing on social networks is inherently social-people exchange ideas, solicit and deliver advice, and develop relationships with other community members (Kozinets, 2002;Mulvey, Padgett, and Lever, 2022).Like others, we contend that crowdsourcing holds enormous potential to identify problems and share solutions in times of crisis.For example, self-help groups and advice on social media may improve affected populations' resilience during a disaster, "replacing their helplessness with dignity, control, as well personal and collective responsibility" (Keim and Noji, 2011).

Methods
Our study examined original messages from Twitter users, a highly inclusive, one-to-many platform where people share information in a relatively synchronous or condensed time frame (Lamberton and Humphreys, 2022).Twitter data gives researchers a window into the volume, sentiment, and expressions of public opinion.We conducted a three-step analysis using data from the Sprout Social COVID-19 database, which holds millions of tweets related to the pandemic.First, we examined key performance metrics of the volume and impact of dignity-related tweets to assess the prevalence of dignityrelated tweets in 2020.Second, we downloaded a sample of Twitter data via Sprout Social and used the Leiden community detection algorithm to detect and study themes in dignity discourse, drawing exemplar tweets to illustrate constituent meanings.Finally, we analyzed themes in the subset of dignity tweets that mention older adults.Ethical approval was waived by the University of Ottawa REB on the basis that data that is publicly available on social media and can be viewed by anyone would not require ethics approval.Our methodology is reproducible and should replicate the reported results, yet a different random sample of tweets may yield slightly different results.
Our cross-national sampling plan aimed to include diverse viewpoints and prevent a single country's tweets from dominating the discourse.Therefore, the sample comprised English-language tweets from countries that posted more than 500 tweets in the 12 months of January to December 2020.Seven countries achieved this level of activity.Also, we decided to prioritize tweets with engagement and followers as opposed to those having little or no impact.So, we sorted the posts in descending order on these metrics before downloading samples proportional to each country's total tweet volume.As a result, the sample includes a total of 1,946 original messages drawn randomly from Twitter, scaled to the United States (1,000 messages), India (300), the United Kingdom (286), Canada (157), South Africa (77), Kenya (65), and Australia (61).
The Leiden community detection algorithm (Traag, Waltman, & Van Eck, 2019) is a state-of-the-art network-based algorithm that can identify clusters of related nodes in a network.In the context of social media data, we use this algorithm to identify groups of concepts that interact with each other in a meaningful way, such as words co-mentioned in sentences and reflecting latent topics or themes.We used the Social Astronomy app (Belanger, 2022) to create a network of words (or n-grams) based on their interactions in social media messages (e.g.words are mentioned together in a tweet).Next, Social Astronomy analyzed the network (or matrix) using the Leiden community detection algorithm to identify clusters of words that interact consistently and cohesively.These clusters may correspond to groups of words that refer to topics or themes.Cluster scores assigned by the algorithm helped retrieve original messages for review by research team members.
The Leiden community detection algorithm results depend on the researcher's selection of resolution parameter setting, which controls the detected communities' granularity and the number of words included in the analysis.A higher resolution parameter will generally detect smaller, more fine-grained communities, while a lower resolution parameter will result in larger, coarser communities.Similarly, including many words will produce a more complex solution than using fewer words.In general, it is helpful to experiment with different values of the resolution parameter to see which one produces the best results given the dataset and analysis objectives.
To improve the accuracy of group membership detection, we ignored stop words (such as "the", "a", "an", and "in") and ubiquitous words, including the search keywords (dignity, #dignity), a quotation acronym (qt), and Sprout Social's COVID-19 database inclusion keywords (covid, covid19, #covid19, #coronavirus, corona, virus, coronavirus, covid-19, pandemic, #covid, #covid_19, #pandemic, #covid19uk).Mindful of our objective to report results with high fidelity to the original data, we iteratively generated solutions using different settings before selecting a comparatively detailed and stable 13-cluster solution that used the top 200 words and a 1.5 resolution parameter setting.

Results
Next, we present the listening insights derived from an aggregate analysis of dignity tweets followed by an analysis of themes and topics that emerge from the messages.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.

Top-level indicators and metrics
The prevalence and engagement with dignity-based tweets The Sprout Social key performance metrics affirm the popularity of dignity in the reporting period of January to December 2020 and establish that the topic was well-established in COVID-era discourse.The keywords (dignity OR #dignity) appeared in 46,420 original messages (not including shared or mentioned), contributed by 41,060 unique authors, generating 2.89 billion impressions and 5.12 million engagements (total likes, dislikes, comments and shares).These values translate to daily averages of 126.8 messages, 112 new authors, 7.9 million impressions, and 13,991 engagements.
Figure A illustrates the volume and engagement levels of tweets from our sample.There are two panels: the top includes all dignity tweets, excluding dignity tweets focused on older adults, depicted in the bottom panel.Comparatively, the volume of dignity tweets focused on older adults is only 4.4% of the total sample.Each circle denotes a tweet, and its size represents the level of message engagement (the total likes, comments, and shares).Most tweets earned low levels of engagement (small dots), whereas some striking exceptions generated exceptionally high engagement (the circles).For reference, the post with the highest level of engagement (=88,390) generated 56,771 likes, 20,174 comments, and 11,445 shares.Notably, Twitter algorithms favour popular posts, amplifying posts with engagement, feeding them into users' streams and generating more impressions.
The heat map shading indicates the relative volume of posts by month (red shading represents the highest density, yellow the least).Some observations: dignity discourse on Twitter was sparse in January and February and rose exponentially in March, April, and May following events including the WHO's pandemic declaration, rising hospitalization and death rates, imposition of lockdown measures, and the Black Lives Matter protests.The volume of tweets decreased and maintained steady levels through the summer and fall months and faded into December.

Tweets with the most engagements
Next, we focus on the posts that earned top-10 engagement scores.Foremost, politicians championed the cause of dignity, led by a handful of well-known members of the Democratic Party, including Hillary Clinton (#1), Bernie Sanders (#4), Chelsea Clinton (#8, #10), and Nancy Pelosi (#9).In addition, his Holiness Pope Francis (#2, #3, also known as Pontifex), medical doctor Craig Spencer (#5), journalist Miles Howard (#6), and author/retired registered nurse Juanita Broaddrick shared views.These posts demonstrated diverse construals of dignity, invoking tensions between democracy and authoritarianism and respect for persons, including older adults, people who are unemployed, patients and healthcare professionals, small business owners, and leaders (not) doing their duty.

Core and peripheral words in tweets
Still, simple word frequency counts and exemplars do not tell the whole story of dignity discourse.Figures B, C, and D include word clouds created using Scimago Graphica that illustrate the relative frequencies of terms, hashtags, and emojis in messages that include the keyword dignity.For example, Figure B shows the prevalence of terms like pandemic and people and negatively charged words such as corruption, authoritarianism, and lawlessness.Figure C indicates how users attach hashtags to messages, linking posts to related conversations, for instance, #blacklivesmatter, #mentalhealth, and #protectreprodignity. Finally, Figure D displays emojis in messages, using visual symbols to convey abstract emotions and feelings, including anger , disgust , and sadness , as well as calls to action to wear a mask , attend to disabled persons , and pray for better outcomes .We also examined mentions of prominent people and brands, signaled by the @ sign, which calls out usernames on Twitter.The list of mentions included the names of many prominent politicians, public health agencies, and news outlets.Increasingly, researchers are attending to hashtags, and emojis as these paratexts that accompany the words may play a prominent role in the reception or interpretation by the public (Bakker, 2022;Luangrath et al., 2022;Völcker, 2020).By extension, these terms, hashtags, emojis, and mentions suggest a bigger picture, like puzzle pieces, yet they do not form a coherent image.

Thematic analysis and word clusters
Next, our attention turned to analyzing clusters of words used together, focusing on the 200 most frequently used words derived using the Leiden community detection algorithm.Terms of 13 groups form clusters or communities of dignitybased topics in our dataset.Our analysis of these results, combined with iterative readings of the verbatim tweets, helped derive the framework illustrated in Figure E, organizing the thirteen clusters into five broad themes.These words and ideas are typical of the cluster, yet variation exists within clusters, and messages can blend ideas from one or more clusters.

Recognize dignity as a fundamental right
The first theme affirms dignity as a fundamental human right (see Figure E for keywords).Users consider dignity a necessary condition for the enjoyment of human rights, as, without dignity, individuals cannot fully exercise or realize their rights.The global pandemic created crisis conditions that eroded a basic sense of humanity and justice as inequalities  in wealth, power, race, and housing conditions threatened individuals' inherent worth and dignity.These inequalities came to the fore with discussions of unequal access to vaccines.For example, one tweet affirmed universal access to vaccines among citizens of South Africa, including migrants.

Uphold the dignity of essential workers
The second theme highlights the importance of upholding the dignity of frontline and healthcare workers.The COVID-19 pandemic threatened the dignity of these essential workers in several ways.One such threat is the inadequate provision of personal protective equipment, which put their health and safety at risk.To address this issue, employers must provide workers with the necessary resources and support to carry out their work safely and effectively, including proper equipment, suitable testing measures, adequate training, and sufficient staffing (The Lancet, 2020;WHO, 2020).
Another way to uphold the dignity of frontline and healthcare workers is by recognizing and valuing their work and contributions.For example, a tweet shared by the Teamsters labour union thanked a member of Local 150 for performing essential work for the Sierra Nevada Memorial Hospital during the pandemic.Employers -and society as a whole-can express gratitude and appreciation for the efforts of these workers by providing them with fair compensation and benefits and offering support and assistance for their physical and mental well-being.
The high stress and isolation many essential workers experienced during the pandemic can lead to burnout, so it is crucial to provide them with the support they need.For example, in a Tweet, the UK-based Living Wage Foundation thanked workers, including school caterers, care workers, cleaners, and delivery drivers, for keeping society going during the pandemic, asserting that they merited a living wage and sufficient work hours to maintain health and live with dignity.
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.By taking these steps, the public can support and empower essential workers and help them continue providing vital services to society.

Preserve the dignity of at-risk populations
The third theme focuses on the close connection between dignity and populations considered vulnerable.For example, India-based WHO South-East Asia issued a tweet that called attention to the need to protect diverse groups in response to the pandemic, including older people, persons with disabilities, people who are homeless, refugees and migrants, people without access to sanitation, and people living in crowded places.Such individuals are more likely to experience disadvantage, discrimination, or exclusion.In addition, due to their limited access to resources, they may struggle with inadequate access to healthcare, education, or social support, which can negatively impact physical and mental healthand undermine dignity and self-worth (Armitage and Nellums, 2020;Croft and Fraser, 2022).Therefore, it is essential to provide at-risk populations with the necessary support to live decent and fulfilling lives.
One factor that can threaten the dignity of seniors, older adults, and elders is ageism.This discrimination and prejudice can manifest as negative stereotypes, lack of recognition and respect, or unequal access to opportunities and services.As a result, older adults may feel marginalized and devalued, undermining their dignity and self-worth.Another factor that can threaten the dignity of older adults is the inadequate provision of healthcare and social support services-especially during the pandemic.For instance, an individual's sarcastic tweet called out the Liberal-National Coalition (LNP) in Australia for failing to protect the elderly supporters who voted them into power.Other messages submitted that when such services are lacking, older adults may face isolation, loneliness, and loss of independence, all of which can harm their dignity.Additionally, lack of access to adequate housing and financial security can threaten older adults' dignity by limiting their ability to meet their basic needs and participate in society.
Similar to older adults, women may also face unique challenges that can threaten their dignity.These challenges may include 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.

Prevent cascading disasters that exacerbate dignity's decline
The fourth theme specifies factors that pose a particular challenge to upholding dignity during the pandemic.Critics describe how well-intentioned COVID mitigation strategies suffer from blindspots of missing information about unresolved social problems, neglecting risks, compounding the harm, and leaving dignity in peril.
First, dysfunctional politics can threaten dignity by promoting discriminatory policies and using propaganda and misinformation.Twitter banter between American political parties epitomizes this theme, as a barbed attack between political rivals generated a blunt counter-response with huge engagement.Dignity was a prominent theme in political discourse in the months leading up to the November presidential election.Political strife can lead to political instability and violence (as demonstrated by the Capitol attack on January 6, 2021), so it is crucial to promote functional and inclusive politics that respect the rights and dignity of all individuals.
Second, food insecurity can undermine individuals' dignity by forcing them to make difficult choices and limiting their access to nutritious and healthy food.The COVID-19 pandemic worsened this problem, leading Nourish Scotland to campaign for communities to respond to food insecurity by providing people with access to healthy food options and ensuring social connections.
Third, poverty can undermine individuals' dignity by limiting their access to basic needs and opportunities.Conversely, addressing poverty can help to restore and uphold individuals' dignity by providing them with the means to meet their basic needs and participate in society.Finally, chronic deficits in many domains require a coordinated solution, as the New York-based advocacy group Women Deliver emphasized, calling for world leaders to embrace international collaboration in recovery plans.
Fourth, healthcare and insurance can support and protect an individual's dignity by providing access to necessary medical care and services and protecting them from financial hardship.For example, an individual from Texas tweeted a reminder that coronavirus does not discriminate and that every single person in the country should be able to see a doctor, irrespective of employment or insurance status.

Attend to death, dignity, and the sanctity of life
The fifth theme considers the consequences of failing to uphold dignity in the face of death and illness.As the COVID-19 pandemic swept across the world, nations faced a tragedy of unprecedented proportions.Each day, the death toll rose as more people fell victim to the deadly virus.Unfortunately, political tensions-particularly in the United States-plus a lack of a cohesive and effective national response exacerbated the situation.As the tragedy unfolded, it became clear that the country failed to adequately protect its people's dignity.An English political pundit posted an updated death toll, ascribed blame, and urged people to vote.
Respecting an individual's dignity during their end-of-life can be crucial to providing decent end-of-life care.Dignity is a quality or state of being worthy of respect and honour, and 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.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.

Tweets focused on older adults
Given our interest in examining older adults' challenges in the pandemic, we next examined the subset of tweets that mention older adults.We flagged messages in the sample that included at least one of the following keyword variants: elder* (n=29), senior* (n=28), older (n=24), pension* (n=11), retire* (n=6), or aged (n=1).This procedure netted 86 unique tweets, equal to 4.4% of the sample.
Most messages echoed the dignity themes discussed previously.Foremost, dignity is a human right protected by the United Nations and World Health Organization, a message affirmed by the UN Secretary-General, who reminded world leaders to respect the rights and dignity of older people.Older persons were disproportionately harmed by COVID-19, as reflected in death tolls and social isolation (Fraser et al., 2020;Sharma, 2021;Wu, 2020).Furthermore, older age intersects with other risk factors, including poverty, disabilities, race, homelessness, immigrants, non-native speakers, and incarcerated persons.
Problems with long-term care homes drew considerable attention as facilities struggled to protect residents from the virus, maintain adequate staffing and care levels, balance social distancing measures, and reduce isolation from family members.The tragic loss of lives sparked calls to hold government officials accountable and to reform the system.
Finally, some seniors' meal delivery services were suspended during the pandemic, placing recipients at risk of not having enough groceries, risking hunger, malnutrition, or even starvation.Poverty increases these risks, as some posts drew attention to low incomes and inadequate pensions or social security benefits.

Discussion
This study examined social media discourse on dignity posted during the first year of the global COVID-19 pandemic.A secondary goal was to explore the use of dignity in conjunction with ageing.Our cross-national sample, Leiden community detection algorithm and thematic analysis of identified clusters revealed five major themes: (a) recognize dignity as a fundamental right, (b) uphold the dignity of essential workers, (c) preserve the dignity of at-risk populations, (d) prevent cascading disasters that exacerbate dignity's decline, and (e) attend to death, dignity, and the sanctity of life.While each of these themes has distinct contributions, the data reveals that essential workers and at-risk populations were identified as groups whose dignity was heavily impacted during the first year of the pandemic.The implications of the five themes are discussed below.
Tweets in the first three themes reinforced the Universal Declaration of Human Rights to promote dignity as a fundamental human right.In line with Mégret and Hoffman's (2009) seminal work, findings demonstrate the importance of contextual factors and intersection with other human rights.For example, women are more likely than men to be frontline workers in the context of the COVID-19 pandemic (Utzet et al., 2022) and, as such, face a greater risk of loss of dignity in the context of healthcare provision during a pandemic.The impact on the dignity and health of essential workers has been raised as a target for policy action (Lancet, 2021).Similarly, the findings of this study align with calls to prioritize the older population, their needs and dignity in times of crisis.Minimizing ageist discourse and maximizing access to healthcare and other services to maintain the health and dignity of this population (Fraser et al., 2020;Lagacé et al., 2020).
Formal and informal communication-via news outlets and social media-have been essential throughout the pandemic for exchanging information and enhancing awareness of risk and public health recommendations (Généreux et al., 2020).However, while modern technologies have immeasurable benefits for providing accessible communication, it is also prudent to consider how media discourse can frame issues and shape beliefs during disasters (Choudhury & Haque, 2018;Wang et al., 2019), including ageist and ableist attitudes (Barth et al., 2021; O'Sullivan and Phillips 2019).Indeed, Fraser et al. (2020) warned of the potential negative impacts of ageist hashtags circulating on social media during the early weeks of the COVID-19 pandemic.Lagacé et al. (2021) report similar concerns about the COVID-19 media discourse labelling all older adults as vulnerable people whom we must "fight for" and not "fight along with."Similarly, Stollznow (2020) describes how prejudice in ageist language perpetuates conflict between boomer and millennial generations, and in the end, impacts everyone.
Our research findings illustrate how social media discourse can be valuable for identifying key issues requiring policy reform.First, the method can amplify the voices of marginalized or underrepresented groups, bringing attention to issues not widely recognized or addressed before.Second, though our analysis was retrospective, the data lends itself to realtime monitoring and rapid response to urgent or pressing issues on social media.Third, it is possible to identify trends and patterns in discussions and debates on social media platforms.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.
The rise of social media has opened new avenues to explore public opinion, yet it is not without 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 understand public views on human dignity fully.Second, while individuals and organizations have control over the messages they create, the algorithms used by platforms significantly influence the extent to which those messages are seen and received by their intended audience (Kozinets and 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.
This study is part of a larger research program focused on how pandemic experiences and exposure to COVID-19 media discourse influenced older adults' perceptions of resilience and vulnerability.It is the first step in a series of research activities where we are exploring how older age and ability are framed in pandemic media discourse.Given the emphasis on dignity in social media, this study provides a glimpse into how the term is generally used in the public sphere and, more specifically, in the context of aging.
As a broader impact, the findings have the potential to inform decision-makers about how the public views dignity and aging.Listening provides feedback on public trust, confidence and priorities and can assist decision-makers by understanding the discourse circulating in social media.Given the conceptualization of dignity as a human right and its relevance for combatting ageism, these findings can help promote healthy ageing throughout pandemic recovery and adaptation to a 'new normal'.

Uğur Gündüz
Faculty of Communication, Prof. Dr., Istanbul University, Istanbul, Turkey The paper entitled "Upholding Dignity During a Pandemic via Twitter" provides valuable insights into how people expressed their views on dignity during the COVID-19 pandemic using social media.By analyzing 1946 original tweets using a combination of text analytics and interpretive methods, the study identifies five major themes related to dignity discourse: recognition of dignity as a fundamental right, upholding the dignity of essential workers, preserving the dignity of at-risk populations, preventing cascading disasters exacerbating dignity's decline, and attending to death dignity and the sanctity of life.This paper contributes to the existing literature by highlighting public sentiment on these themes and suggesting how such insights might inform policies and practices to protect and promote dignity.The paper covers an important research area, and overall, the paper is well articulated and presented.However, I would like to see the following concerns addressed: So, I recommend minor revision as indicated below: 1-The paper would benefit from a more detailed explanation of the selection criteria for the tweets included in the study.Specifically, clarity is needed on how the tweets were deemed relevant to the study's themes of dignity, and any exclusion criteria should be explicitly stated.2-A brief discussion on the limitations of the interpretive methods used and how they might impact the study's findings would strengthen the research's credibility.3-While the identified themes are compelling, the discussion would benefit from a deeper exploration of how these themes interact with each other.For example, how does the recognition of dignity as a fundamental right influence actions towards at-risk populations?Expanding on these connections could offer more nuanced insights into the public discourse around dignity.4-The study mentions a cross-national sample but does not delve into regional variations in dignity discourse.Given the global impact of the COVID-19 pandemic, highlighting any geographic differences in the perception of dignity on Twitter could enrich the analysis.If data allows, a comparison between countries or regions might reveal culturally specific concerns or expressions of dignity.5-The conclusion section could be expanded to offer more concrete suggestions for future research directions based on the findings.Furthermore, providing specific policy recommendations or interventions to address the issues raised by the study would make the manuscript more impactful.For instance, suggesting ways social media platforms could promote dignity in public discourse during crises could be a valuable addition.6-There are minor typographical and grammatical errors that need correction to improve the manuscript's clarity and professionalism.Good luck with your research !..

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility?Partly

Are the conclusions drawn adequately supported by the results? Partly
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: New Media, Social Networking, Media and Cultural Studies I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
Reviewer Report 08 March 2024 https://doi.org/10.5256/f1000research.142542.r187199© 2024 Kietzmann J.This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Jan Kietzmann
Peter B. Gustavson School of Business, University of Victoria, Victoria, BC, Canada I really enjoyed reading this piece.As a social media researcher, I was mostly interested in the methodology for using Twitter data, as I have seen many different approaches.Here, I was thoroughly impressed by the rigor demonstrated by the authors.In terms of the topic itself, I have to admit that I am completely new to the focus on dignity.I have researched online exploitation of people, in particular exploitations of a sexual nature related to children.In my work, I had not considered the dignity literature, and can now see how much value this angle would add to my own work.Thank you, to all of the authors, for introducing this important lens.It is without a doubt that other readers will learn lots, too.I understand that their focus was on COVID related comments but can see how their approach can be replicated for many other instances, far beyond exogenous shocks like COVID.
In terms of readability, I would also like to compliment the authors on the attention they paid to ensure that a newbie could easily consume and understand this deep and important topic.Likewise, the methods section was so clear that others, interested in running a Twitter study themselves, could now replicate the methods employed.
As far as Twitter data is concerned, the platform comes with the usual pros and cons, and I was happy to see that the authors acknowledged this.As a non-native speaker, I am keenly aware of the dominance of English on Twitter and social media in general.After having lived and/or spent a lot of time in many of the countries that were included, I was curious if one should try to correct for the cultural and linguistic differences between the countries included.I am not convinced that this would strictly speaking be necessary, but the authors could add a short reference to other articles that have used similar geographically diverse settings etc.I was looking forward to the discussion, and once again, was impressed with how all of the pieces were connected here.Well, done.The one point I noted was that I was not clear on who the main audience of this piece was.Is it public policy makers?If so, the focus can be sharpened from the beginning to the end.If the audience is broader, maybe a simple table that outlines what different segments can take away from this study would be interesting.Please do not see this as a strong point of criticism, but as food for thought.
As you can see, I really, really enjoyed this piece and have nothing that I think I'd change.It's a narrowly focused, rigorously executed and superbly documented research study.As a complete aside, I enjoyed the emoji word cloud -very interesting.I will now have to learn how to interpret emojis in a rigorous way :-).

If applicable, is the statistical analysis and its interpretation appropriate? Not applicable
Are all the source data underlying the results available to ensure full reproducibility?

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Social media research I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
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.

○
The acronym LGBTQI+ is well-known, however I do not know what 2S stand for, perhaps it would be better to spelled it out.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'?
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.
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.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'.

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

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).
Consider adding a 'conclusion' section.

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.

○
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.

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.

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

○
(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.

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

Results:
A short summary of the main results which relates to the research questions (as well as ○ methodological focus) would be helpful.
I particularly like the final paragraph of page 5.The tone is pleasantly authoritative.
○ (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?○ (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?○ (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?
○ (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.
○ (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.
○ (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).
○ (P.9, para 8) Needs a reference or several references for 'critics describe…' ○ (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.

○
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.
○ (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.
○ (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).

○
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.The start to this section is strong, and the tone is appropriate.The summary of the main themes is particularly helpful.
This article has an appealing focus on the articulation of dignity through Twitter during the start of the COVID-19 pandemic.It touches on key concepts around dignity and multiple disadvantage which were exacerbated during the pandemic.It would benefit from some restructuring (particularly in the results section, see comments) and a deeper critical engagement with concepts of dignity, gender, age, and the wider political and crossnational social contexts within which the texts studied were created.It would also benefit from a clearer explanation of the necessity of this study, and of the trends and oppositions in existing theoretical and empirical literature in this research area.

○
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com

Figure A .
Figure A. Sample of dignity tweets by month with engagements.

Figure B .
Figure B. Word cloud of terms in dignity tweets.

Figure C .
Figure C. Word cloud of hashtags in dignity tweets.

Figure
Figure D. Word cloud of emojis in dignity tweets.

Figure
Figure E. Dignity themes and word clusters.

○
Figures and Tables:I find that figures and tables are clear, easily interpreted, and well-formatted.
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, Please see for instance: Stommel W, Rijk L de.Ethical approval: none sought.How discourse analysts report ethical issues around publicly available online data.

○
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.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.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.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.
○ (P.○I understand ○ I'm unclear from the introduction what the exact aims or hypotheses of this study/article are.Can these be stated explicitly?○ ○ Methods:(Para 2, p. 4) I disagree that Twitter is a highly inclusive platform.What about people who have literacy issues or with a lack of digital skills, internet access, or other unmet access needs which prevent them engaging with it?