Social Media Use in the Face of Disaster

An Exploration of Communication Practices Among Stakeholders Affected by Typhoon Haiyan in the Philippines

Abstract

This study focuses on the functions played by social media when Typhoon Haiyan, one of the strongest storms ever recorded on Earth, pummeled the Philippines in November 2013. In this context, traditional communication channels become non-operational, and other non-traditional information sources and communication platforms, such as Facebook and Twitter, became salient. This study examines patterns of social media use by various groups—government officials, journalists, and citizens—affected during the disaster. Through interviews with 18 respondents, the study found that in terms of social media, Facebook became useful across the three groups which used the social media site in a variety of ways—for disseminating, informing, mobilizing, and even coping. The analysis also found that individual and structural level factors shaped the ways with which different people used social media during the disaster.

Introduction

On November 8, 2013, Typhoon Haiyan (aka Yolanda), the strongest tropical storm ever recorded on the planet, plummeted parts of the Philippines, leaving massive destruction mostly from giant sea waves blown by the strong wind that swept communities, killing between 6,000 and 10,000 people. The typhoon tore through Tacloban City, in the province of Leyte, and affected about 13 million people (Chan, Liu, & Hung, 2013)1. The Philippines, an archipelago, is composed of more than 7,000 islands, and the communities affected by the devastation are interspersed in islands separated by bodies of water, making mediated communications necessary but also fragile during disasters. The powerful storm knocked down power and phone lines, and the geography made communication in the affected areas extremely difficult. Some 10 percent of Filipinos also live outside the country, and many overseas Filipinos had to rely on social media to monitor the situation in their home communities and keep in touch with their loved ones.

Social media use is particularly relevant in the context of the Philippines. The Philippines ranks among the top countries in terms of social media use. In 2012, approximately 30 million people (about one third of the country’s population) were Facebook users. [1]2 In the context of typhoon Haiyan, some anecdotal evidence also suggests that social media played an important role among affected populations and government officials. [2]3

Despite the considerable destruction, and the geographic, economic, and logistical constraints, the initial response to the disaster, both internally as well as from the international community, demonstrated a superior level of coordination compared to previous similar disasters (Cranmer & Biddinger, 2014)4. This professionalization of disaster response shows encouraging signs thanks in part to improvements in organizational and technological innovations. Nevertheless, there are still several obstacles in the long road to recovery, which includes access to sanitation services that can exacerbate health problems. Some of these problems can be partially attributed to a lack of coordination between response agencies (Chiu, 2013)5.

This study focuses on the ways social media was used in this natural disaster, as well as the factors that shaped these particular uses. We also develop best practices of social media use under these types of scenarios that are applicable to contexts outside the Philippines. In the next sections, we shall review relevant literature; present the research questions, the method, and the results; and finally discuss potential applications.

Theoretical Considerations

The use of social media has in recent years received some attention from researchers studying natural disasters and human responses to such crises. Using social media has become a ubiquitous activity among a large population of individuals around the world, therefore having a significant potential in serving various functions before, during, and after natural disasters. These events, such as the devastating effects of Typhoon Haiyan, Hurricanes Sandy and Katrina in the U.S., and the nuclear crisis in Fukushima in Japan, highlight the importance of social media in uncertain situations. Indeed, “online media discourse is an important venue in the social construction of risk” (Binder, 2012, p. 275)6.

Current studies of social media use during disasters take mostly an organizational perspective (Lundgren & McMakin, 2013)7. These studies focus on the ways in which organizations are utilizing social media, the limitations they face (e.g. time, money), and proposals for best practices (e.g. Dabner, 20128; Feldpausch-Parker, Parker, & Peterson, 2012)9. For example, Freberg and colleagues (2013)10 tested the effectiveness of social media messages sent by organizations during Hurricane Irene, identifying the following relevant functions: communicate quickly, be credible, be accurate, be simple, be complete, and communicate broadly.

On the other hand, there are fewer studies looking at the use of social media by individuals such as lay populations, decision makers, and journalists. For example, some research suggests that adoption and continued use of social media after a disaster is significant as these media are perceived by individuals as a way to help in the response and recovery efforts, as well as in rebuilding a sense of community ( Hughes & Palen, 2009)11. However, not much research has been conducted in this respect focusing before and during the event.

In addition, it is important to consider structural factors such as level of Internet penetration, level of social media use, and the use of these media from a geographical perspective. Indeed, the Philippines has a large number of social media users, and also a large number of Filipinos living outside the country. Starbird and Palen (2010)12 examined retweets by people outside the area of emergency, and they found their use of media information to be significantly higher than that of local people.

As media audiences evolve with recent technological changes, they are no longer passive receivers of information. Thus, in this study, we analyze the uses of social media before, during, and after Typhoon Haiyan in the Philippines, and how these uses intersect by focusing on the experiences of journalists who covered the disaster, the people who were affected by it, and the individuals that had to take decisions to deal with the problem.

Research Questions

RQ1. How did stakeholders directly and indirectly affected by Typhoon Haiyan use social media?

We focused specifically on government officials who had to make decisions to deal with the disaster, as well as local reporters who covered the issue, and citizens who were active in reporting the issue through social media.

RQ2. What factors shaped these patterns of use during the crisis?

For RQ2, we specifically differentiated between individual (internal) and structural (external) factors that influenced individuals’ use of social media.

Methods

Data collection

We conducted fieldwork in Tacloban City in the Philippines in March 2014. We conducted 18 in-depth interviews with government officials (n=5), residents (n=9), and local reporters (n=4). Tacloban City is one of the several areas hit by the storm. Since it was a bustling city prior to the tragedy, and relatively more crowded than the other areas, the storm had a significant impact in Tacloban.

We followed a snowball sampling procedure to select interviewees. We first coordinated with a local journalist who managed to introduce us to a few key respondents in the beginning of the fieldwork. We also searched social media sites for people who had posted information about Haiyan prior to, during, and immediately after the tragedy, and we tried to contact them. These initial set of respondents later referred us to other people who were active on social media during the tragedy. Through this process, we managed to interview respondents from three different groups: individuals at local governmental institutions with decision-making capacity during the disaster; residents who used social media during the disaster; and journalists based in Tacloban City who covered the disaster.

Data processing and analysis

The interviews were conducted in Filipino (one of the researchers is Filipino) and transcribed by local transcribers. The transcripts were then analyzed using a thematic and inductive approach. No preconceived theoretical constructs were specifically used to code, but the researchers coded for themes that were considered instances of social media use, and the explanations of the reasons they relied on those communication channels. The results presented in the next section follow the broad themes identified in the analysis.

Results

The first research question asked about the different ways various stakeholders used social media during Typhoon Haiyan. In our fieldwork, we focused on three groups of stakeholders: affected residents, the local government, and the local media. This section documenting the patterns of social media use in Tacloban City during the storm is divided into these three groups of stakeholders.

Local Government

Disseminating. Several units of the city government maintained Facebook accounts, some of which were used to disseminate information prior to the storm. The chief security officer of the city at that time, who also headed the barangay (village) affairs unit which oversaw village officials, said “information was properly disseminated” prior to the storm through various channels, including Facebook. “We have many pages on Facebook which we use for information dissemination,” he said. For example, the barangay affairs unit had its own Facebook page that was supposedly being monitored by the village officials. An officer from the regional civil defense office (OCD) who was based in Tacloban City said that prior to the storm, he and his co-administrator of the regional OCD’s Facebook page posted weather advisories to inform their followers about the approaching storm. But the issue with this use is that it perpetuates a top-down approach to communication, with the presumption that crucial information disseminated on Facebook—such as announcements and evacuation plans—were reaching residents. There appeared to be no mechanism for the city government to ascertain whether particular channels of information dissemination, such as social media, were working effectively.

Coordinating. Facebook also served as a quick and efficient communication platform for government officials to coordinate and check on one another. The regional officer for the OCD we interviewed had a personal Facebook account that he did not use for information dissemination about the storm, as he considered it his personal space. But he used his personal account to communicate with his superiors at the national office based in Metro Manila. “Since I was based in the office, instead of texting, we just used Facebook because it was faster.” He also used the office’s Facebook page to communicate with local government units (LGUs) in the region. Those maintaining the Facebook accounts for the LGU offices under the regional office “were required to reply or comment” to the posts on the regional account. This was one way to ensure that the messages posted on the Facebook page of the regional office were being read. Commenting was considered a form of acknowledgement that a piece of information shared by the regional office on Facebook was attended to by the LGU offices. However, this also represents a linear communication pattern. The information shared was also usually limited to factual advisories, such as weather bulletins.

Journalists

Monitoring. The journalists we interviewed used social media, particularly Facebook, to monitor the path of the storm prior to landfall. For example, a newspaper reporter said that ordinarily, he uses Facebook to monitor what others in his network are posting that might offer leads for potential stories. This was also true during the days leading to Haiyan. He said he was monitoring Facebook in the early morning of November 8, 2013, coming across posts from his Facebook friends about the typhoon’s path.

Disseminating. The journalists also used social media to disseminate information they thought would be helpful to others. In a way, they were extending their journalistic reach through social media. For example, an online reporter we interviewed said she posted useful tips on her Facebook page, such as encouraging people to prepare emergency kits. She posted tips she found after conducting some Google searches, aggregating information from a variety of sources, such as the country’s weather bureau. She felt that it was part of her role as a journalist to do so, even in her own personal social space: “Personally, I feel that we are accountable to others. They have a right to know these things. Maybe it is because I feel like I understand what the storm surge is about more than the average person.” She was referring to the unexpected storm surge that sent walls of water crashing into neighborhoods, killing thousands. She also posted updates early in the morning, with her last post being about the storm making a landfall in the town of Guian, about two hours southeast from Tacloban City. But aside from posting factual information, she also expressed some of her personal emotions on Facebook. In one of her last few posts that Friday morning, she said: “Lord, please spare us.” She said she posted this status “so that other people would feel that we really needed divine help.”

Informing. When Haiyan finally hit Tacloban City, it unleashed so much power the city, nor the country, has never seen before. Power, phone, and internet lines were cut, isolating the affected areas from the rest of the world. The following day, the social welfare and development department (DSWD) managed to offer free internet access in a makeshift office in front of the city hall, which was directly facing the sea but was located on top of a hill and was therefore spared from the storm surge. News about the free internet access spread slowly among the residents, many of whom were stranded in areas isolated by debris. Two of the journalists we interviewed were among the earliest people to show up. The newspaper reporter said it took him more than an hour of waiting to finally access his Facebook account. Six laptops, based on his recollection, were made available. When he opened his account, some 20 people had already left messages on his Facebook wall—relatives from Manila and outside the country, friends, and even editors from Manila—all checking if he was okay. He said that each person who showed up in the city hall grounds was allowed only three minutes to use a laptop—a short span of time meant to ensure that as many people as possible could use the free service. “Chatting was not allowed—just post your message, that’s it. I just said ‘don’t worry, we’re okay.’”

The online reporter we interviewed also walked to the city hall with her husband, who was also a stringer for a national newspaper. There, they were able to charge their phones and laptops. They also managed to quickly send reports to their editorial desks using the free internet service. While her husband was using one of the available laptops to quickly send a story, she was also in the other laptop. “I opened my Facebook to tell everyone I know, my friends, my family, that we are okay. That we are alive, although we lost everything.”

The newspaper reporter actually came back the following day—but his main purpose was still to inform his family and friends he was okay. We asked him about his work as a journalist. But as a resident, whose house was also swept by the storm surge, he was also a victim. “I was immobilized for three to four days,” he said, citing the disruption in traditional communication lines. “And I couldn’t go around to see what happened,” he added. His example demonstrates how journalists themselves are not immune to disasters. Social media would have provided a way for him to report to his superiors, but the problem was more than just access. Himself a victim of the disaster, he also had to deal with his own trauma.

Recovering. The TV reporter we interviewed was deployed near the sea prior to the storm’s landfall. The warning of a storm surge meant her news organization wanted a team to monitor the situation near the waters. This only exposed her and her two cameramen to danger. They lost their cameras to the storm surge, and she was left taking photos and videos using her phone. She managed to submit one video to her TV station days later, but for a couple of weeks, she refused to upload on Facebook any of the photos she took. “I remembered the photos after a few weeks,” she said. “I thought that I’m still lucky enough to have survived.” This was when she decided to upload photos of the devastation she had witnessed that also documented the perils of being a reporter during a disaster. The photos she uploaded got “positive feedback,” with her wall filled with friends expressing well-wishes and gratitude that she survived. “It made me feel happy to share [the photos].”

Citizens

Disseminating. Some citizens used social media to pass on information they came across from other sources about the approaching storm. For example, a freelance photographer, who owns a restaurant in the city, posted information about Typhoon Haiyan days before it made landfall. Himself a blogger, he found Facebook a useful tool to reach more people. “It was obvious [that] some people just didn’t know where to get information,” he said. Thus, he aggregated information from other sources, such as the website of the country’s weather bureau, “basically to warn people.”

Informing. In this study, we consider dissemination as limited to passing on factual information external to the user. In contrast, informing others refer to providing information about oneself. For example, a college student we interviewed said that as soon as she got internet access when her family managed to get on a plane to the province of Cebu a few days after the storm, the first thing she did was post this status update on her Facebook: “We’re all alive. We’re all okay.” Her professor, whom we also interviewed, also said she used Facebook to inform others that she survived: “I think everybody was looking for me, I am sure, so I posted on Facebook.” She did this in both her two accounts—her personal, and the one she uses to keep in touch with her students. “I said something like, ‘Praise God for this second life.’”

Information-seeking. Some residents also used Facebook to seek information about people they know in the absence of any information. The college student we interviewed said when she opened her Facebook a few days after the storm hit, she found her account tagged by other friends, asking where she was and if she was okay. A court employee said that when phone lines were restored, at least in a few areas, the first thing she did was use Facebook on her phone to keep track of people she knew. It was through Facebook that she found out that some of her friends were missing. One resident we spoke with also recalled how one of her cousins posted on Facebook the names of their relatives and where they were. “It was as if every person was itemized,” she said, “to ensure that when [relatives and friends] read the post, they would know [their loved ones] are safe.”

Mobilizing. We also interviewed a project manager for a new non-profit organization mobilized to help residents from the nearby town of Tanauan. She was in Tacloban City during our fieldwork. Her organization was born out of information-seeking amid scarcity of information right after the storm. Former residents of the town based in Metro Manila, worried about their families and relatives who remained in Tanauan, decided to create a Facebook page to ask not only for information about their loved ones, but also to solicit donations. Facebook turned out to be an effective platform for donations—the members of the non-profit group posted names of the donors and photos of their donations on Facebook, which also served as an effective tool for documentation.

The local chapter of the Philippine National Red Cross in Tacloban City also used Facebook to invite more donors and volunteers in the aftermath of the storm. “We posted our activities on Facebook, our relief activities, the purpose is to encourage other donors,” a Red Cross volunteer said. “So we were like marketing Red Cross at the same time by posting pictures on our Facebook account.”

Complaining. Some residents used Facebook to complain about the inefficiency and slowness of relief operation efforts. None of our respondents explicitly claimed to have used Facebook to air complaints. But the college student we interviewed said she had been tagged on Facebook in status updates that expressed disappointment and anger toward the snail-paced relief efforts. The restaurant owner we spoke with also recalled seeing a lot of negative messages on Facebook, but he said he decided to post only positive things. “If you post negative things, what will happen to us?” he remembered himself thinking. Instead, he said: “I got to provide inspiration, and motivate people to stand up.”

Coping. Social media also became a platform for many residents to face their pain and come to terms with what they experienced. For example, the court employee we interviewed referred to the process of posting photos and looking at them later as a form of “stress debriefing.” Though her family was fortunate for being spared from the typhoon’s wrath, she had taken videos of water seeping into her house, a reminder of the experience. Watching the videos weeks later helped her children laugh, remembering that they also saw some fish left in the house when the flood receded.

The Red Cross volunteer we spoke with also talked about posting photos of a pool at the Leyte Park Hotel weeks after the typhoon. “I posted that to reminisce about the past, but I also said, “This is not an experience worth missing.”

Factors Shaping Use

The second research question centers around the factors that have shaped the usage patterns documented in our interviews. Our analysis classified two clusters of factors that have shaped social media use during Typhoon Haiyan in the city of Tacloban.

Individual-level factors. This cluster refers to personal characteristics and circumstances that shaped how stakeholders used social media in the face of disaster. Some residents were more social media savvy than others. Some used Facebook more often than others. The newspaper journalist we spoke with used Facebook in general to monitor others, instead of self-expression, and this was also true with his use of Facebook during the typhoon. He did not share information or tips about the storm, unlike the TV and online reporters we talked to. Instead, he used it to monitor what was happening, and after the storm, used it to communicate with friends, superiors, and family to inform them he was safe. The online reporter we spoke with clearly saw sharing factual information about the storm’s path as consistent with her role as a journalist, and she saw Facebook as another platform to perform that role during the typhoon.

Though we see that many of the people we interviewed found novel uses of Facebook in the face of disaster—in effect it functioned as a platform for one-to-many communication, as they were constrained with limited time and access, that communicating individually with multiple people was impossible—how they regarded Facebook as a platform prior to the disaster shaped the extent to which they used it during Haiyan. Some kept on sharing information, others uploaded photos after. But these were behavior they have engaged in on Facebook even before the storm. These personal qualities and circumstances also shaped the extent to which they used Facebook during Haiyan.

Structural factors. The extent of devastation wrought by Typhoon Haiyan and the technological affordances of Facebook can explain why many residents, and even the government, turned to Facebook right after the storm to inform and update people from outside the city about their conditions. Power, phone, and internet lines were cut and took several days to restore. The news media teams deployed in the affected areas also suffered from damaged communication equipment that hampered communications. The government and the private sector eventually offered free phone calls to the residents a few days after the storm. But what emerged from our interviews is the important role Facebook played after the typhoon.

Facebook is very popular in the Philippines, a country that has been described as the social media capital of the world. Internet penetration remains low, but a local mobile phone carrier partnered with Facebook to offer unlimited Facebook access to prepaid customers for about six months, starting a month before Haiyan. Thus, many of the residents we spoke with who talked about sharing information about the storm were mostly accessing Facebook from their mobile phones. This also explains Facebook popularity over other social media sites, such as Twitter and Tumblr, during the disaster.

This omnipresence of Facebook also intersects with Facebook’s ability to allow users to engage into a one-to-many communication, a form of what Manuel Castells called as mass self-communication. Thus, right after the storm, when the city was isolated from the rest of the world, residents turned to Facebook to reach out to their family and friends, especially in a country marked by a strong collectivistic culture. Their Facebook networks already composed of peers and relatives, residents in Tacloban City turned to Facebook amid limited access to communication. Instead of calling people individually, residents chose to post a quick status update on Facebook to inform their loved ones about their conditions. Some said they were okay. Some asked for help. It was a quick and efficient way to communicate during a period when communication lines were down. The residents found a useful function for the status update feature during a crucial time.

We saw that the use of Facebook also evolved in the different stages of the disaster. Some users shared factual information about the storm, presumably based on the news coverage the storm attracted locally and internationally, as Haiyan made headlines when it was spotted outside the Philippines packing a force never seen before on the planet. But the people of Tacloban City have been used to storms, and the country gets an average of 20 storms per year. Weather forecasters mentioned the possibility of storm surge—but it was a term almost alien to residents, government officials, and even journalists. This explains the lukewarm response to, and some confusion about, the impending storm. This was evident in the limited use of Facebook prior to the storm.

The extent to the devastation also made it impossible to use social media—dependent on internet access and devices that run on electricity—during the storm. Power, phone, and internet lines were cut about an hour after Haiyan made landfall in Tacloban. Communication, however, is crucial especially in relief and rescue operations. Thus, when the government managed to offer free internet access the day after the storm, many residents who heard about the service walked for hours, depending on how far they lived from the city hall, amid debris and sights of hundreds of bodies, young and old, men and women, lying on the streets. Faced with limited access, Facebook proved useful to many residents to update their relatives about their plight, as well as to ask for help.

Applications of the results

The results presented above can be used to improve communication practices using social media channels during a disaster. Based on our fieldwork, we developed the following recommendations and applications:

The government should integrate a social media-based communication infrastructure in emergency preparedness plans. Our interviews revealed that after the typhoon, users had limited access to internet connection. This access was offered by government offices, and made available at the city hall. We recommend that government officials, in future disaster-preparedness plans, should carefully plan access points for such services in anticipation of communication lines being damaged. They should clearly and widely communicate where these locations would be, and what time periods they will be made available, prior to the disaster, so that residents can also plan ahead.

National and local governments of affected areas should develop official social media channels. The government officials we interviewed explained that social media channels were useful in communication various issues related to the extent of the damage and relief efforts. However, there was no clear strategic or coordinated use of these communications channels within and across each institution involved with the disaster. We propose the need for government institutions to develop social media emergency plans that serve as the first source of reliable information by individuals affected by natural disasters.

News organizations should also establish procedural guidelines when reporting life-threating disasters. Our interviews with journalists highlighted some of the difficulties they faced during disaster coverage. Journalists used social media to disseminate information, monitor the situation, and communicate with their acquaintances. However, during these times, the lines between being a journalist and a victim of the disaster became blurred. How far should the reporters go when reporting the issue when their own lives or those of their loved ones are in danger? This is a conversation that should happen in newsrooms prior to such dangerous assignments. Communication lines should be planned ahead, with contingencies laid out in the event of power, phone, and internet breakdown.

Citizens should be educated about the variety of alternative communication lines, such as social media, that they can access in the aftermath of a disaster that compromises traditional channels. What happened during Typhoon Haiyan demonstrates how media users quickly learn to adapt technologies to suit their peculiar needs during a communication network paralysis. While this is admirable—a testament to human ingenuity—people will be more empowered if they are equipped with more information about communication alternatives prior to such disaster. Educating residents about the value of social media channels will make social media use during disasters more efficient.

The results of this study can also have applications to other contexts outside the Philippines. Despite the limitations and constraints faced by stakeholders in the use of social media (e.g. internet access) during disasters, people in the Philippines rely heavily on this form of communication. The residents of Tacloban City have demonstrated the potential of these media in connecting various stakeholders, aiding in rescue and reconstruction efforts, and monitoring the situation and even in collective coping, among others. This study highlights the possibilities of extending social media use through innovative partnerships, especially in other developing countries where internet access remains to be non-universal. The reason Facebook played such an important role in Tacloban City during Typhoon Haiyan is that a national mobile phone carrier introduced a temporary free Facebook service with selected prepaid plans. The service lasted for six months, but the short-lived service offered a peak into the promise of such arrangements especially during times of crises. Such free service can be offered in anticipation of peculiar communication needs during natural disasters through synergies between governments and businesses.


  1. Chan, E. Y., Liu, S., & Hung, K. K. (2013). Typhoon Haiyan and beyond. The Lancet, 382(9908), 1873. 

  2. http://www.internetworldstats.com/asia.htm#ph 

  3. http://www.nbcnews.com/news/other/typhoon-haiyan-filipinos-use-social-media-ensure-no-victim-goes-f2D11581515 

  4. Cranmer, H. H., & Biddinger, P. D. (2014). Typhoon haiyan and the professionalization of disaster response. New England journal of medicine, 370(13), 1185-1187. 

  5. Chiu, Y.-T. (2013). Typhoon Haiyan: Philippines faces long road to recovery. The Lancet, 382(9906), 1691-1692. 

  6. Binder, A. R. (2012). Figuring out #Fukushima: An initial look at functions and content of US Twitter commentary about nuclear risk. Environmental Communication: A Journal of Nature and Culture, 6(2), 268-277. doi: 10.1080/17524032.2012.672442 

  7. Lundgren, R. E., & McMakin, A. H. (2013). Risk communication: A handbook for communicating environmental, safety, and health risks: Wiley. com. 

  8. Dabner, N. (2012). ‘Breaking Ground’ in the use of social media: A case study of a university earthquake response to inform educational design with Facebook. The Internet and Higher Education, 15(1), 69-78. 

  9. Feldpausch-Parker, A. M., Parker, I. D., & Peterson, T. R. (2012). 350.org: A Case Study of an International Web-based Environmental Campaign. In A. Carvalho & T. R. Peterson (Eds.), Climate Change Politics: Communication and Public Engagement (pp. 211-242). Amherst, NY: Cambria Press. 

  10. Freberg, K., Saling, K., Vidoloff, K. G., & Eosco, G. (2013). Using value modeling to evaluate social media messages: The case of Hurricane Irene. Public Relations Review. 

  11. Hughes, A. L., & Palen, L. (2009). Twitter adoption and use in mass convergence and emergency events. International Journal of Emergency Management, 6(3), 248-260. 

  12. Starbird, K., & Palen, L. (2010). Pass It On?: Retweeting in Mass Emergency. Paper presented at the Proceedings of the 7th International ISCRAM Conference–Seattle. 

Suggested Citation: Takahashi, Bruno and Edson C. Tandoc. Social Media Use in the Face of Disaster: An Exploration of Communication Practices Among Stakeholders Affected by Typhoon Haiyan in the Philippines. Quick Response Grant Report series; 251. 2014.