The Coronavirus (COVID-19) is the latest and most widespread pandemic affecting the entire world. As of August 21, 2020, more than 188 countries and territories report more than 22.5 million cases, resulting in about 800 000 deaths. Governments and policymakers have made concerted efforts to minimize the ongoing COVID-19 pandemic in their countries.Reference Boccia, Ricciardi and Ioannidis 1 –Reference Van Bavel, Baicker and Boggio 5
Countries have been severely affected by the COVID-19 disaster and developed disease policies to cope with the virus.Reference da Economia 6 –Reference Wenham, Smith and Morgan 8 For example, the Brazilian government established the “War Budget” Constitutional Amendment (PEC) that allows the separation of expenses incurred to combat COVID-19 from the budget of the Federal Government, creating an extraordinary regime to allow the expansion of public expenditures during the ongoing crisis, without the constitutional barriers that currently restrict federal spending.Reference da Economia 6
Scholars also have tried to understand how COVID-19 exerts a serious impact on human life and environments.Reference Brodin 10 –Reference López-Feldman, Chávez and Vélez 12 For instance, Burki reports that as of April 14, Latin America registers more than 65 000 cases of COVID-19. Ecuador is badly affected, with reports of corpses left abandoned on the streets.Reference López-Feldman, Chávez and Vélez 12
Latin American countries, especially, have been seriously affected by the ongoing COVID-19 pandemic (see Figure 1). For example, as of August 21, 2020, Brazil ranks second with about 3.5 million cases and 112 304 deaths, and Peru places sixth with 567 059 total cases and 27 034 deaths.
Figure 1. Total confirmed COVID-19 cases as of August 23, 2020.
Source: https://ourworldindata.org/grapher/total-covid-cases-region
However, prior studies have barely highlighted who plays an important role in Social Network Services (SNS) to get valuable information on COVID-19 in Latin America.Reference Meneses-Navarro, Freyermuth-Enciso and Pelcastre-Villafuerte 13 – 15 Understanding SNS for COVID-19 would provide significant implications for governments and centers for disease control to provide relevant information on COVID-19 to the public in a timely manner to minimize COVID-19 damage and infection.
In this sense, this study aims to highlight who plays important roles in the information networks of COVID-19 in Latin America. For the objective, this study explores SNS for COVID-19 in Latin America by employing social network analysis (SNA) for Twitter, which is one of the most popular SNS. This study tests 3 hypotheses as follows:
H1. Latin American countries would show different social networks because the impact of COVID-19 is differentiated by countries.
H2. Key players would be different from country to country because the countries’ social, political, and health care conditions vary.
H3. Social networks for COVID-19 among groups in the countries would demonstrate different characteristics because the countries have various information systems and networks.
This study selects 6 countries in Latin America: Brazil, Peru, Colombia, Chile, Argentina, and Ecuador, in descending order of the number of COVID-19 patients (Table 1). To the best of my knowledge, this study is the first article exploring SNS for COVID-19 by employing SNA across Latina American countries.
Table 1. COVID-19 In Latin America
Source: Google news
Literature Review
COVID-19 is a new infectious disease caused by a new strain of coronavirus. COVID-19 was unknown before the outbreak began in Wuhan, China, in December 2019. World Health Organization published a comprehensive package of guidance documents for countries, covering topics related to the management of an outbreak of a new disease on January 10, 2020. The organization declared a public health emergency of international concern on January 30, 2020. The situation continues to evolve rapidly in China and internationally, and further spread of the virus cannot be anticipated.Reference Heymann and Shindo 16
Governments and scholars have investigated how COVID-19 affects human health and countries.Reference Pan, Liu and Wang 17 –Reference Bedford, Enria and Giesecke 19 Latin American countries, especially, are rapidly developing COVID-19 policies to provide technical and financial support to successfully respond to COVID-19. 20
Among Latin American countries, Brazil, Peru, Colombia, Chile, Argentina, and Ecuador are some of the most seriously affected by COVID-19. For instance, Brazil ranks second, Peru places sixth, Colombia takes eighth, Chile ranks ninth, Argentina places 13th, and Ecuador takes 26th for the number of COVID-19 patients across the world. This study specifically reviews each country’s COVID-19 situations and literature as follows:
Brazil has one of the best health systems in Latin America, whereas capacity is highly uneven across regions. The spread of the virus toward lower health care capacity in poorer areas, such as the North and Northeast of Brazil, poses a challenge to Brazil’s ability to respond to COVID-19.Reference Bautista 21
Peru shows an interesting situation because it is one of the countries most affected by COVID-19, despite responding with the most drastic measures in Latin America and implementing the biggest fiscal stimulus package in the region (17% of GDP). This is due to income vulnerability, poor living conditions, low-quality health systems, lack of social systems to protect the most vulnerable, and weak social contract.Reference Amariles, Granados and Ceballos 22
Colombia shows an increasing concern both for mortality and the limited capacity of the health system to respond efficiently to the needs of COVID-19 patients, despite the implementation of the rules established by the Colombian government. Colombia shows a case fatality rate among COVID-19 patients of 0.6%. 23
Chile declared a national health emergency for beginning Phase 1 of the epidemic in the country on February 8, 2020 to cope with COVID-19.Reference Villalobos Dintrans, Browne Salas and Madero-Cabib 24 While the Chilean government enacted some policies to protect vulnerable people from COVID-19, the government should provide a wider perspective to suggest better health care systems considering financial scarcity, access to health services, mental health issues, and long-term care.Reference Gemelli 25
Argentina also experienced the severity of the COVID-19 pandemic. Although the Argentine government implemented an early quarantine, COVID-19 had caused more than 329 040 infections and around 67 300 deaths as of August 22, 2020. The first case was confirmed on March 3, 2020 and the first death was reported on March 9, 2020. As the number of cases ascended in Argentina and around the world, the Argentine government implemented national quarantine on March 19, 2020 to forbid free circulation of people, cancel flights, and seal borders.Reference Kirby 26
Ecuador struggles with a huge outbreak in its main city, Guayaquil, with miserable cases of many dead bodies being reported from houses some days after the person’s death. The Ecuadorian Government reported that 6700 people died in the Guayas province in the first 2 weeks of April, far more than the usual 1000 deaths during the same period in previous years. There are reports across authorities of Ecuador struggling to enforce the lockdown.Reference Lopez and Gallemore 27
On the other hand, some scholars employed SNA for SNS to understand information networks of COVID-19.Reference Pascual-Ferrá, Alperstein and Barnett 28 –Reference Calderon, Fisher and Hemsley 30 For example, Yum highlights that President Trump plays the most important role in social networks for COVID-19 in the United States.Reference Calderon, Fisher and Hemsley 30 Pascual-Ferrá et al. show that the network of conversations around COVID-19 is highly decentralized, fragmented, and loosely connected in the COVID-19 Public Discourse on Twitter.Reference Yum 29 However, they have not highlighted the information networks for COVID-19 for Latin countries. Therefore, this study explores who plays an important role in information networks for COVID-19 across Brazil, Peru, Colombia, Chile, Argentina, and Ecuador.
Research Methodology
This study employs SNA to highlight public key players for COVID-19 in Latin America. SNA explores the connections among social units and the outcomes related to these connections.Reference Uddin, Hossain and Wigand 31 SNA has been widely used to explore networks and their participants by calculating networks of actors.Reference Alotaibi, Mehmood and Katib 32 This study utilizes Twitter data to explore social networks systems for COVID-19 across public key players. Twitter has been widely used for big data analyses.Reference Arslan, Birturk and Djumabaev 33 –Reference Dunne 34 This study observes Twitter data stream between August 6 and August 13, 2020 based on the keywords COVID-19 and the countries and employs the NodeXL program every day to secure the best data for COVID-19 networks. While this study employs the NodeXL program during the period, the program collects the data from December 2019. This study chooses the best data set for the analyses (August 11 and August 12). This is because the data set has the highest number of Twitter users, communication networks, and suitable contents for COVID-19.
This study employs NodeXL to highlight the social networks of Latin America for COVID-19. NodeXL is a visualization software program, which supports social networks and content analysis. NodeXL has been extensively utilized to highlight social network systems in academic fields.Reference Langville and Meyer 35 This study employs PageRank to capture public key players for Twitter users (see the explanation for PageRank as follows). This study chooses the top 10 key players among all Twitter users based on the magnitude of the PageRank (see Table 2 for descriptive statistics).
Table 2. Descriptive statistics
This study first shows PageRank, in-degree centrality, and out-degree centrality to highlight social networks for COVID 19 in Latin America. PageRank is an algorithm used by Google Search and is a way of measuring the importance of nodes based on counting the number and quality of links to the node. For example, if a Latin key player receives more high-quality links of COVID-19 networks, the Latin key player is considered as more important. PageRank has become the successful social network model due to its virtual immunity to spamming, its query-independence, and Google’s search model.Reference Ding 36 PageRank can show which Latin key players play an important role in social networks of COVID-19 by counting the number and quality of links to each Twitter user.
In-degree centrality calculates the number of networks that point inward at a node. For example, if a Latin key player is mentioned 100 times by other people in the COVID-19 networks, the key player’s in-degree centrality metric is 100. In-degree centrality can reveal the Latin key players that function as central hubs in COVID-19 networks.
Out-degree measures the number of networks that originate at a node and point outward to other nodes.46 For example, if a Latin key player mentions 100 other people in the COVID-19 networks, his out-degree centrality is 100. Out-degree centrality can demonstrate Latin key players that have strong level of engagement with other members of the COVID-19 networks.
Results
Figures 2-4 demonstrate that social networks for COVID-19 show different characteristics according to centrality types and countries. Tables 3-8 highlight the top 10 key players according to countries. In the Brazil networks, the most significant characteristic is that Brazil has no Latin key players. This is the biggest difference between Brazil and other countries because other countries have many Latin key players. To be specific, Al Jazeera English, which is the news channel owned by the Al Jazeera Media Network, ranks first. The Associated Press, which is the American not-for-profit news agency, places second. Reuters, which is the international news organization, takes third. BBC News World, which is the British Broadcasting Corporation International channel, ranks fourth. CNN International, which is the American Cable News Network International, places fifth. Overall, news media plays an essential role in the social networks of Brazil. They place from the first to the eighth.
Figure 2. PageRank.
Figure 3. In-degree centrality.
Figure 4. Out-degree centrality.
Table 3. The top 10 key players in Brazil
Note. LA: Label, PG: page rank, L: Latin key player, W: world key player
Table 4. The top 10 key players in Peru
Table 5. The top 10 key players in Colombia
Table 6. The top 10 key players in Chile
Table 7. The top 10 key players in Argentina
Table 8. The top 10 key players in Ecuador
In the Peru networks, the Ministries play a significant role in social networks for COVID-19. For instance, Ministerio de Salud, which is the Ministry of Health, ranks first, and Consejo de Ministros, which is the Council of Ministers, places ninth. Another important characteristic is that RT channels, which are international television networks, play an important role in social networks of Peru. For instance, RT en Español, which is the first RT TV channel in Spanish, takes third, and RT Última Hora, which is the latest news of international events, places sixth. Also, Peruvian news media exert a crucial impact on social networks of COVID-19. For instance, Diario La República, which reports on Twitter what is happening in Peru and the world, ranks second. El Comercio, which is the Peruvian newspaper based in Lima, takes fourth. Diario Gestión, which is the Peruvian economy and business newspaper, places fifth.
In the Colombia networks, the most important findings are that they have the highest number of Latin key players (8) (with Ecuador) and all Latin key players are Colombian key players. Latin key players show the highest rank from the first to the sixth. For example, Elespectador, which is the newspaper with national circulation within Colombia, ranks first. Noticias Caracol, which is the latest national and international news in Colombia, places second. BluRadio Colombia, which is the radio station in Bogotá, Colombia, takes third. MinSaludCol, which is the Ministry of Health and Social Protection of Colombia, ranks fourth. Iván Duque, President of the Republic of Colombia, takes fifth. EL TIEMPO, the main news from Colombia, places sixth.
In the Chile networks, diverse key players exert a significant impact on social networks of COVID-19. For example, Andrés Allamand, the Chilean politician and the founder and one of the past leaders of Renovación Nacional, places first. EFE Noticias ranks third. DW Español, which is the regional version of official German TV Deutsche Welle for the Americas, takes fourth. Alerta News 24, alerts from all over the world 24 hours a day in real time, ranks eighth. Diario La República, reporting on Twitter what is happening in Peru and the world, takes tenth.
In the Argentina networks, governmental key players exert a pivotal impact on the social networks. To be specific, the president plays the most important role in social networks of COVID-19. For example, Casa Rosada, the office of the President of Argentina, places first, and Alberto Fernández, the president of Argentina, ranks second. The Ministry of Health of Argentina has a crucial effect on social networks. Ministerio de Salud, the Ministry of Health of Argentina, places fifth. Latin key players from other countries also play an important role in social networks. For example, Quebrando o Tabu, which is the storytelling organization focused on raising awareness and engagement around human rights, democracy, and development in Brazil, places sixth. Con el Mazo Dando, the Venezuelan television program, ranks seventh. El Universal, which is the Colombian newspaper, takes eighth.
In the Ecuador networks, they show some unique characteristics. Rafael Correa, the former president of Ecuador, plays the most significant role in social networks of COVID-19, whereas Lenín Moreno, the current president of Ecuador, does not rank within the top 10 key players. Also, they show the highest number of Latin key players (8) (with Colombia). For instance, El Universo, one of the largest daily newspapers in Ecuador, ranks second, and Ecuavisa Noticias, the national leader in news, places fourth. El Telégrafo Ecuador, which is the Spanish-language daily newspaper in Guayaquil, Ecuador, takes fifth, and Ecuavisa, which is the Ecuadorian free-to-air television network, ranks sixth.
Next, this study employs cluster analysis by utilizing the Clauset–Newman–Moore cluster algorithm (see Figure 5 and 6). Cluster analysis is a methodology for the task of assigning a set of objects into groups so that the objects in the same cluster are more similar to each other than those in other clusters. The Clauset-Newman-Moore algorithm is a fast-hierarchical agglomeration approach because many social networks are sparse and hierarchical.Reference Kuchler, Russel and Stroebel 37
Figure 5. Social networks for the typical case.
Figure 5 shows the social networks of COVID-19 for the typical case. In the Brazil networks, nodes show a dense circle, and some nodes are distributed in the right side of the circle. Key players show the most concentrated pattern among 6 countries. In the Peru networks, nodes create some circular layers, and many nodes are located in the outside of layers. Some key players, such as Global Citizen (W3) and Con el Mazo Dando (L5), have some distance from other key players. In the Colombia networks, nodes are located as a large circle, and some nodes are dispersed in the right and under side of the circle. All key players tend to be placed in the central part except for NTN24 Venezuela (L7). In the Chile networks, nodes demonstrate a large circle, and some nodes are connected with the circle from the southwest part. In the Argentina networks, nodes tend to be placed as an ellipse, and some nodes are around it. Key players are apt to be located in the central part. In the Ecuador networks, nodes show the most dispersed patterns. Nodes exhibit a large circle, small circle, and some networks in the right side of the circle. All key players are located in the central part of the left side of large circle.
Figure 6 shows the social networks of COVID-19 according to groups. In the Brazil networks, there is no key player in the largest group (group 1). Instead, Al Jazeera English (W1) plays an essential role in group 2, and Reuters (W3) exerts a significant impact on group 3. All key players have their independent group except for BBC News World and UNHCR (United Nations High Commissioner for Refugees) (group 6). Brazil shows the most closed communication networks across groups among 6 countries.
Figure 6. Social networks according to groups.
In the Peru networks, governmental key players play an important role in the largest group. For example, Ministerio de Salud (L1) and Consejo de Ministros (L7) plays a pivotal role in group 1. Also, El Comercio (L3), Diario Gestión (L4), and RPP Noticias (L6) exert a crucial impact on group 2. Group 2 shows not only the highest number of key players, but also the most active communication networks across groups. Key players are in large groups except for W3 (group 21).
In the Colombia networks, there is no key player in the largest group. MinSaludCol (L4) plays a crucial role in group 2. Iván Duque (L5) and Gustavo Petro (L8) have a significant effect on group 3. All key players have their independent group except for L5 and L8. Colombia shows the most open communication networks across groups among 6 countries.
In the Chile networks, there is no key player in the largest group. BioBioChile (L3), Cooperativa (L4), and Ministerio de Salud (L6) play a significant role in group 2. Group 1 shows active communication networks with other groups. Other key players except for L3, L4, and L6 have their independent group.
In the Argentina networks, Alberto Fernández (L2) and Todo Noticias (L3) play an important role in the largest group. Casa Rosada (L1) and Ministerio de Salud de la Nación (L4) have a significant impact on group 2. Other key players have their own group. Group 1 shows dynamic communication networks with other groups.
In the Ecuador networks, many key players exert an important impact on the largest group. Five of the top 10 key players belong to group 1. For example, El Universo (L2), Ecuavisa Noticias (L3), Ecuavisa (L5), Salud_Ec (L7), and El Comercio (L8) have a significant effect on the group. Also, other key players are located in large groups.
Table 9 shows the top tweeters in the social networks in Latin America. The top tweeters are differentiated by countries, but some countries have the same tweeters in the social networks. For example, Brazil has pulpnews, and Peru and Colombia have caraotadigital as the first tweeter. Chile and Argentina have extravzla, and Ecuador has la_patilla as the top tweeter. For the sake of readers, Pulpnews is the news media from the US that delivers fast crime news. CaraotaDigital is the news media from Venezuela that delivers national events, politics, and entertainment. Extravzla is the news media from Venezuela that delivers a variety of news. La Patilla is the Venezuelan news website that is one of the top 600 websites visited in the world and among the top visited in Venezuela. All Latin countries have the Venezuelan top tweeter except for Brazil.
Table 9. The top tweeters in the social networks
Table 10 highlights the common top tweeters across Latin America. Extravzla has the highest number (5), meaning that it ranks within top 10 for 5 of 6 countries. Noticias24 and sumariumcom show the highest number, as well. Noticias24 is the news media from Venezuela for Latin America and the world. Sumariumcom is the Venezuela-based website aimed at global readers who want to be aware of what is happening in Latin America. The results show that Venezuelan news media play a significant role in Twitter for COVID-19 in Latin America.
Table 10. The common top tweeters in Latin America
Discussion
COVID-19 is one of the worst pandemics across the world in the human history. Developing COVID-19 policies and minimizing the damages of COVID-19 are some of the most important tasks for governments and policymakers to protect their citizens. In this background, understanding social networks of COVID-19 would be one of the most effective ways to release important COVID-19 information and resources to the public in a timely manner. Therefore, this study explores who plays an important role for information networks of COVID-19 in Latin America by employing SNA for Twitter.
This study finds several significant results, which are consistent with prior studies as follows: firstly, Latin countries show different social networks for COVID-19 because various factors, such as cultural and social factors, in Latin America influence these networks, which is a similar finding with prior studies.Reference Weiss 38 –Reference Biswas, Majumder and Dawn 40 For example, Weiss highlights that digital and social media in Latin America, specifically in Argentina, Brazil, Colombia, Mexico, and Peru, play a different role in daily news gathering and reporting routines in their countries.Reference Yum 39
Secondly, different key players play an important role in social networks for COVID-19 across countries, which concurs with previous literature.Reference Forster and Heinzel 41 –Reference Azzaoui, Singh and Park 43 For example, Rufai and Bunce highlight that G7 world leaders play a different role in response to the COVID-19 pandemic in their countries.Reference Azzaoui, Singh and Park 43
Thirdly, SNS, such as Twitter, Instagram, YouTube, Facebook, and TikTok, can play an essential role in social networks for COVID-19, which is consistent with prior articles.Reference Masciantonio, Bourguignon and Bouchat 44 –Reference Hansen, Shneiderman and Smith 45 For instance, Kuchler et al. highlight that online social network data can be useful to governments and disaster experts hoping to forecast the spread of COVID-19 based on Facebook.Reference Weiss 38
This study has some limitations as follows: firstly, this study explores who plays an important role in information networks of COVID-19 in Latin America only based on Twitter. Future studies should explore the key players of information networks for COVID-19 by utilizing various flatforms, such as Facebook, Instagram, and YouTube. Secondly, this study only explores 6 Latin countries, whereas there are many countries in Latin America. Other scholars should investigate information networks for COVID-19 in other Latin countries, which are not highlighted in this study. Thirdly, this study highlights information networks for COVID-19 based on PageRank, in-degree centrality, and out-degree centrality, while there are many centrality methodologies. Future studies should employ various centrality methodologies.
Conclusions
COVID-19 is one of the most urgent issues for governments to address in protecting their citizens. COVID-19 has been one of the worst pandemics and has seriously damaged people’s lives and economic growth. Therefore, understanding social network systems of people would be a good strategy for governments to release important information on COVID-19 to the public as soon as possible. In this sense, this study explores who plays an important role in COVID-19 information by employing SNA for Twitter in Latin America.
This study provides some important findings as follows: firstly, key players in Latin America play various roles in COVID-19 social networks, differing from a country to a country. For example, Brazil has no Latin key players, whereas Colombia and Ecuador have 8 Latin key players in the top 10 key players. Secondly, the role of governmental key players also varies across different countries. For instance, Peru, Chile, Argentina, and Ecuador have the governmental key player as the top key player, whereas Brazil and Colombia have the news media key player as the first. Thirdly, each country shows different social networks according to groups. For instance, Colombia exhibits the most open social networks among groups, whereas Brazil shows the most closed social networks among the 6 Latin countries. Fourthly, several top tweeters are common across the 6 Latin American countries. For example, Peru and Colombia have caraotadigital (Venezuelan news media) and Chile and Argentina have extravzla (Venezuelan news media) as top tweeters.
This study suggests important implications as follows: firstly, governments should explore their key players for COVID-19 because each country shows different characteristics of social network systems. Secondly, countries and policymakers should highlight important Latin key players and governmental key players to deliver relevant information on COVID-19 to people in a timely manner. Thirdly, governments and centers for disease control should analyze the characteristics of social networks for COVID-19 according to groups because they have different network systems. Fourthly, Latin American countries should cope with the COVID-19 pandemic together because they have common key players and top tweeters, and because COVID-19 is a global issue that spreads across nations.
