The Yolanda Tragedy: 7 Lessons in Early Emergency Response
November 20, 2013
By Eric Aseo
Last month, when the 7.2 earthquake struck the Philippine provinces of Cebu and Bohol, I was in the southern city of Zamboanga facilitating dialogues between Muslim and Christian leaders to alleviate possible religious tension following the September siege that displaced thousands and threatened the good relationship of the city’s two faith communities. It was the furthest thing from my mind that an even more devastating disaster would happen just a month later, right in Tacloban City, where I had left my wife and kids in safety (or so I thought) and in the province of Eastern Samar where I grew up playing in the gentle edges of the mighty Pacific Ocean. Typhoon Yolanda (internationally known as Haiyan) will forever change my idea of safety.
On November 8, Typhoon Yolanda (known internationally as Haiyan) struck central Philippines, particularly the eastern coasts of the islands of Leyte and Samar. Photo/Eric Aseo
At around 4 am on November 8, Friday, the day Typhoon Yolanda hit Samar and Leyte, my brother called me in Manila from Eastern Samar asking me to check the latest weather bulletin. Electricity in the province is already gone, he said; the wind was getting stronger and it was blowing low, threatening to sweep away most houses in our small town. By 6 am, I called up my son in Tacloban City who in his deadpan way said, there isn’t much rain and the wind wasn’t as strong as expected. That calmed me down, but not for long. When I called again at around 7 am, I could no longer contact them and the communication gap went on for two and a half more days. Information coming from the government and the media was bleak and sketchy.
The absence of reliable information took its toll on the victims’ families who were away when Yolanda struck. A friend working in Saudi Arabia said that in the period he’d been trying to reach his family, he couldn’t eat, he couldn’t sleep, and he couldn’t work. The social media was also replete with pleas from overseas Filipino workers for information on their families. Lesson No.1: Immediately restoring communication lines after a disaster or putting up disaster-proof communication systems would save a lot of people from worries and would keep them productive even in the face of disaster.
On the morning of November 10, Sunday, I tried my luck with military cargo planes, since all flights to Tacloban were cancelled. But I ended up helping soldiers at the Villamor Airbase explain to other waiting civilians, some of them hysterical, that flying in rescue teams and equipment first was more important than flying us home. At the base, I also had to absorb the grief of some who received early bad news from home. Lesson No. 2: Entry to and arrival points from disaster areas should have personnel who are trained to deliver clear messages and to provide psychosocial services to waiting families and arriving victims. This will prepare families for the worst and provide early interventions to victims with trauma.
After taking on those roles for a while and waiting for another half a day, I decided to take the 24-hour route by land to Tacloban. A friend on his way to see his mother offered me a ride. We had similar plans – to locate our families and pull them out of the city immediately. We traveled non-stop, stopping only in Naga City where my friend got an elated call from relatives who said they had managed to find his mom and she was fine. For 10 hours I prayed to get the same call. It was only in Calbayog City, four hours away from Tacloban on a regular day, when I finally got word from my wife. They were home and safe. That was 10:35 am, November 11, Monday.
While still in Calbayog City somebody tipped us off to load up on fuel as it was running out even in Catbalogan, another small city two hours away from Tacloban. We filled the vehicle tank full and brought another 20 liters of fuel, enough to take us in and out of Tacloban. We were warned that food and water were also in short supply in Tacloban and people were coming to Catbalogan to replenish supplies, putting pressure on the city’s own supply. Lesson No. 3: After a disaster, or even before, it makes sense to beef up the supply of food, fuel, and water in areas that will not be hit. In all likelihood, they will serve as supply points for the disaster-affected areas.
In the aftermath of Typhoon Yolanda, basic services, such as access to fuel, were cut off due to the destruction. Photo/Eric Aseo
When we reached Catbalogan, we tried to gather all the information we needed to reach Tacloban safely. Some told us to use only motorcycles but sans the backpacks; it’s the fastest and safest way into the city, they said. Others warned us of overloading our vehicle with food and water; it’s an open invitation to being mobbed. A few suggested that we leave our vehicle in Catbalogan, take a motorized boat to Tacloban instead, and sneak out our families by sea. We noted all these bits of advice, but decided to continue on our journey, as planned, in our vehicle, carrying all our supplies.
At around 4 pm, exactly 24 hours after we left Manila, we crossed the 1.3 mile-long San Juanico Bridge that connects the islands of Samar and Leyte. We needed to travel 12 kilometers more to get into the heart of the city. This did not prove easy. Piles of debris obstructed parts of the road along with some cadavers. There were also vendors hawking items like milk, fuel, and shampoo, as well as an endless procession of vehicles, evacuees, and looters carrying sacks of rice, trays of eggs, and chickens. It took us eight hours to cover the 12-kilometer stretch. Once in a while a pair of policemen or soldiers would pass us seemingly unaware of the mayhem; other commuters, including rescue workers, from time to time would get off their vehicles to check the prices of looted items sold on the roadside. Lesson No. 4: Road clearing should be a priority. Roads that are not cleared delay relief and rescue, prolong the victims’ and their families’ agony, and even encourage mob rule.
At around midnight, we finally reached Tacloban City. It was drizzling and the city was in pitch-darkness. I asked to be dropped at the tent of a police colonel who headed a 30-man team and requested to wait at their outpost until daybreak. His men stopped people coming out of the city with sacks – some of which revealed items obviously taken from ransacked stores and warehouses. The policemen were kind enough to offer coffee and allow these people to take food items they would need for three days. But the policemen also warned them not to repeat what they did and asked them to tell their communities such acts are not tolerated even during disasters.
While sipping coffee with some of the policemen, two officers and a local businessman who owns a nearby warehouse came. The officers organized an eight-man team to secure a wealthy enclave of Tacloban. The businessman told us four of the five suppliers of prime commodities in the city had already gone and were not coming back any time soon. The biggest rice miller, he said, would also be leaving the following day. His story strengthened my resolve to pull out my family as soon as I could reach them.
What set me thinking, though, was the judiciousness of deploying eight policemen to secure the houses of the rich. If that pattern of deployment would continue, I thought, they won’t max out the services of the 30-man contingent. Lesson No. 5: Security managers should exercise prudence in the deployment of limited forces. In communities not badly affected by the disaster, unarmed multipliers perhaps can be organized to protect their own communities freeing up more policemen and soldiers to lead in recovery, maintain law and order, and guard crucial establishments.
At around 5:30 am, November 12, Tuesday, I left the police outpost and finally began walking toward home. Around ten meters away from the tent, I noticed more than 30 cadavers covered with mats and blankets. I pushed on, covering my nose with a handkerchief. On the right side of the Maharlika Highway, facing the town of Palo in Leyte province, most communities appeared not to have been as badly affected by the typhoon as I expected. These communities could have been sources of volunteers in the relief efforts if they were assured of the security of their homes and families. Lesson No. 6: Even typhoons as devastating as Yolanda often spare pockets of communities who can fill in the scarcity of volunteers. Relief organizations should find ways to tap these local resources.
As I walked on I met entire families, young men and women, walking toward San Juanico Bridge where buses headed to the Samar provinces were waiting. Some were students going back to their home towns. A few hopped in when a school bus sent by the city government of Calbayog stopped. Others were walking toward the Tacloban port. I later learned that two island municipalities of Samar sent motorized boats to bring their people back home. Lesson No. 7: Nearby local governments not badly affected by the disaster should ferry home temporary residents of the affected area. This will decongest the area and lessen competition for the limited food and water supply.
I arrived home at around 7 am and after five hours, I was back on the road with my wife and kids. We watched the evening news about Tacloban in Calbayog City. Cebu was hosting some of the victims and help from other countries was pouring in quickly. I finally had a chance to read and reply to text messages. Priests and ulama friends from Zamboanga City offered to pray for my family’s safety. They’re OK and out of danger now, I said. Deogracias, Alhamdullilah, they replied. Who says we’re not connected?
Eric Aseo is a program officer for The Asia Foundation in the Philippines. He can be reached at email@example.com. The views and opinions expressed here are those of the individual author and not necessarily those of The Asia Foundation.
Related programs:Environmental Resilience
Related topics:Disaster Management, International Development, Typhoon Haiyan
Introduction: Anti-social behavior and self-preservation are often assumed to be normal responses to threats and disasters; on the contrary, decades of research and empirical studies in social sciences showed that pro-social behaviors are frequently common and that solidarity is the typical response to a variety of threats. The main objective of this study is to investigate and describe survivors’ behavior, especially solidarity, according to the presence of familiar persons and to the perception of physical danger, elaborating the framework of Mawson’s social attachment theory.
Methods: In order to investigate these relationships, a behavioral research was carried out involving 288 people affected by the December 8th 2013 Haiyan Typhoon (Yolanda).
Results: Results revealed that solidarity was predominant and people reacted collectively and actively taking part in relief activities. Furthermore, we found strong solidarity and help towards strangers and unfamiliar persons.
Discussion: Investigating how people react is essential to develop a more efficient and effective response strategy, especially in the immediate aftermath of a disaster when disaster managers have little control of the situation and people rely on themselves; the natural tendency to help others can be essential to reduce losses and to fill the temporal gap between the event and the arrival of the organized relief unit.
Many myths and incorrect beliefs are common about disasters; anti-social and irrational behaviors are usually assumed to be the most common response to danger and disasters1,2,3,4,5 panic, for example, is frequently and erroneously used to describe disaster response6,7, despite it is a rare phenomenon that occurs only in presence of specific affective/cognitive factors8. Decades of research and empirical assessment on disasters and emergencies revealed that what really happens after a disaster frequently differs from what the conventional wisdom would suggest: victims tend to respond effectively and creatively9 and the behavior in such circumstances is very meaningful and far from many conceptions of irrationality7,8. Uncertainty and confusion characterize the disaster scenario but the fear felt during the early phase of the emergency, instead of leading to maladaptive behavior, turns in altruistic acts with people start looking for ways to secure their own and others safety10. Survivors’ decision to behave, adapt and respond in the immediate aftermath of a disaster is anticipated by an extended period when people mill about11 and starts with the assessment of a changed social and physic environment12. Within this process, the individuals’ assessment of personal risk (or the perception of) as variables of environment perception is one of the main factor influencing the decision 12,13,14. More, due to the fact that people are used to move in group12,13,14,15,16, survivors’ behavior cannot be merely explained as sum of behaviors of individuals who influence each other. Survivors can be with familiar people, strangers and a combination of both. In any case, people use to maintain groups, by seeking the proximity of familiar people, or to create a new relationship with strangers who share the same feeling and fate. Shared identity, in fact, resulting from the sharing of the same emergency experience, enhances the expression of solidarity16, mutual aid and sociality8,17,18,19,20. There are many studies on people helping familiars persons in different emergencies and disaster situations9,10,12,16; in 2005, Mawson proposed the “social attachment model” according to which seeking the proximity of familiar persons and places rather than fleeing is the typical response to a variety of threats20. This theory describes the response to threats and disasters as a movement towards a situation that is perceived as safe (but not necessarily objectively safe). According to Mawson’s discussion, combining the factors of perceived physical danger (called precipitating condition) and location of attachment figures (called predisposing condition) results in a four typology behavioral model that encompasses a wide spectrum of collective responses to threat and disasters (Fig. 1).
This paper was criticized by some authors, especially regarding the undefined nature of the attachment behavior21 and the lack of empirical research and new contributions to the current theory of social behavior in disaster22. Moreover, this theory doesn’t describe and explain the bonds that rise during a disaster among people that are not familiar or known. Although the assumptions of the social attachment model have been strongly criticized, the framework of Mawson’s model should be held useful to describe ideal responses patterns or collective actions. In this paper we investigate and analyzed survivors’ behavior in the immediate aftermath of a disaster; we focus on solidarity and mutual help and we studied how social variables, such as the presence of familiar people or strangers, and environmental aspects, such as the perception of danger influence people’s response. The main aim is to understand and describe the actual behavior of survivors in emergency and disasters in order to help disaster planners and managers to develop better response strategies.
In this paper we designed a case study survey23 with the purpose of investigating the behavior of victims in the aftermath of a real event that was carried out after the Haiyan (Yolanda) typhoon. The typhoon (category 5) hit and devastated portions of Southeastern Asia, more specifically several regions in the Eastern cost of the Philippines. According to the National Disaster Risk Reduction and Management Council24 of the Republic of Philippines, the event killed at least 6,300 and caused 28.689 injured people and 1601 missing persons. The relief team operated with many difficulties and this forced people to react in the immediate aftermath without a specific coordination and support of relief teams. The study area is San Roque (Fig. 2), a seaside barangay (municipality) in Tanauan (Leyte), one of the most affected communes of the Eastern Visayas Region (VIII), where Municipal Local Government reported 1,200 people dead and almost 2,000 missing, out of a population of almost 50,000.
A cross-sectional survey was conducted because it is able to provide highly viable and excellent data on behavioral attitude of people after disasters25. Between December 11th and December 16th 2013, 288 volunteers, who experienced the Typhoon Haiyan in first person, were asked to fill a structured questionnaire. Given the chaotic and challenging scene which prevails in the aftermath of disasters and the difficulty to trace a demographic profile of the population because the lack of available data, the sample was selected according to people’s ready availability to be recruited26. Considering that the questionnaire delivery mode also has to be tailored on the so called “survey-taking climate” (e.g. social context in which the research is being undertaken), it was opted for the adoption of a mixed method which integrated face-to face contact with self-completion of the questionnaire27. The contact with people who experienced the event occurred in two moments: a public meeting in the meeting hall of the city and during the distribution of the relief supplies in the town hall. The participants were supported by investigators in order to provide an in-depth insight into the purposes of this research and to support them in the completion of the questionnaire in an anonymous way, reducing the possibility to obtain a low return rate due to misunderstandings or social desirability bias. The questionnaires were administered in Tagalog, an Austronesian language spoken as first language by a quarter of the population in the Philippines and as second language by the majority of people28. An alternative English version was available. Data collection and analysis were carried out following the policies adopted by the Università Politecnica delle Marche (UNIVPM). Each responder provided verbal informed consent prior to participation and voluntarily decided whether to participate or not. The authors collected no identifying participant data. All the interviewees were informed about the scientific purpose of the study survey and on the use of data collected.
Starting from the four-fold typology of responses proposed by Mawson (Fig. 1), a structured questionnaire was developed in order to test the spectrum of collective responses to threats in presence of “predisposing conditions” (shortened with PRED), that describe the proximity of attachment figures, and “precipitating conditions” (shortened with PREC), that quantify the perception of the physical danger (and not the degree of perceived danger used by Mawson). The attachment figures were defined for participant as familiars, relatives, friends or even known people. A structured format containing dichotomous questions was chosen in order to obtain answers as much objective as possible; except for the question related to PRED and PREC, the questionnaire contained only polar questions. After a set of questions, aimed at evaluating the demographic profile, respondents were asked to express their perception of danger (mild/severe) and whether they were in presence or not (present/absent) of attachment figures. Furthermore, they answered to a set of questions aimed at investigating their behavior in the aftermath of the Typhoon. More specifically, respondents indicated if they decided to leave instead of stay and wait for the relief, if they relied upon a person who appeared to be a leader, if they interact with strangers and unknown people, if they engaged in a response behavior when dealing with the situation, if they took part in rescue activities, and finally if they tried to collect additional information on the event. The questions were not directly correlated one to each other, but they intentionally followed a sequence that aimed at addressing the decision-making process over the time. The relationship between PRED and PREC conditions and behavioral questions was explored by performing Chi-square tests for independence with Bonferroni correction for multiple comparisons. According to Mawson’s four-fold typology response, the sample was divided in behavioral profiles and the association between each profile and behavioral questions was investigated by applying Chi-square tests for independence with Bonferroni correction for multiple comparisons. Eventually, results were combined into an alternative profile chart that describes victims’ behavior in response to threat and disaster.
Respondents were averagely 37.3 years old (SD ± 16.1) with 10 years of median education level. The majority of them were females (62%). Ninety-one per cent of the respondents already experienced an extreme weather event. There is a fine balance between respondents who were with attachment figures (55%) in the aftermath of the event and those who were alone or with strangers (45%). More than three out of four of them perceived a severe level of physical danger (76%). Overall, most of the respondents (82%) tended to stay on the site instead of leave it with a narrow majority of them (51%) engaged in an information seeking behavior. Sixty-three percent of the respondents claimed to have relied upon a person that appeared as a leader and 69% of them reported to adopt a response behavior after the Typhoon. Despite the majority of interviewed people (59%) took part in rescue activities, less than half of them interacted with strangers (47%). The associations of PRED and PREC with each of the behavioral variables were analyzed either singularly or in the four-fold form, which derived from the combinations of the PRED (present/absent) and PREC (mild/severe) outcomes. Chi-square tests for the independence (Yates continuity correction) showed the same significant association only for the variable “having relied upon a leader” (PRED – χ2 (1) = 8.26, p<.005; PREC – χ2 (1) = 4.73 p<.05). Table 1 reports the cross-tabulation between each behavioral question and the two conditions (predisposing and precipitating). Bonferroni correction (α level = 0.008) was applied in order to control the familywise error rate associated with the multiple comparisons performed in the Chi-square test.
*significant at 0.008 level (Bonferroni correction)
As evincible by results, people were more likely to rely upon a leader in presence of attachment figures while perceiving severe physical danger. Considering the PRED, people were more likely to adopt an information seeking behavior in absence of attachment figures (χ2 (1) = 19.62, p<.001), while they were more likely to engage in response behavior (χ2 (1) = 16.45, p<.001) and to participate in rescue activities (χ2 (1) = 10.16, p<.005) in presence of attachment figures. With regard to the PREC, people were more likely to leave the scene when perceiving mild physical danger (χ2 (1) = 15.71, p<.001), and to interact with strangers when they perceived severe physical danger (χ2 (1) = 9.39, p<.005). Once cross-tabulated with the behavioral variables and combined to obtain the four-fold response typology, PREC and PRED showed the results reported in Table 1. A further set of Chi-square tests for independence had been performed to evaluate the associations between the attitude of the respondents towards each of the behavioral variables and the response profile they belong to. Results of the analysis are shown in Table 2.
*significant at 0.008 level (Bonferroni correction)
Statistically significant associations were found between four-fold profiles and the following behavioral questions: decision to stay on site in the aftermath of the event (χ2 (3) = 30.64, p<.001), with only Profile A exceeding the 50%; the information seeking behavior (χ2 (3) = 31.5, p<.001), with only Profiles B and D exceeding the 50%; the confidence in a leader (χ2 (3) = 15.7, p<.005), with only the Profile B below the 50% threshold; the engagement in an response behavior (χ2 (3) = 18.53, p<.001), with again only the Profile B under the 50% threshold; the interaction with strangers (χ2 (3) = 13.34, p<.01), with Profile D the unique over the 50%; and the participation in rescue activities (χ2 (3) = 50.24, p<.001), with Profiles A and B below the 50% threshold. The results of behavioral questions for each of the Profiles were plotted in a Kiviat diagram in order to provide a displayable and more intuitive version (Fig. 3).
Survivors’ behavior in the aftermath of a disaster is mostly characterized by solidarity and collaboration within groups (pre-existing and emergent) instead of anti-social behavior and self-preservation. Such behaviors are influenced by many factors but the presence of attachment figures and the perception of danger play a pivotal role in decision-making process and reaction to emergency and disaster. Results showed that the most common response is to stay on site; only a part of the individuals in presence of attachment figures and with a perception of low physical danger left the scene after they have recreated the pre-existing group and when all the members were safe. The confidence on a person that drives the decisions of the rest of the group (leader) predominates in presence of attachment figures, in which pre-existing norms, roles and bonds keep working and driving the decision-making process of the whole group. However, the overall behavior was active and collaborative, especially for people in presence of attachment figures while the perception of danger seemed not to be a considerable driver. Solidarity and collaboration were strong and directed even to strangers and unknown people, expressing the same stressful feeling and trying to solve the situation together. The decision to take part in rescue activities was strongly influenced by the presence of familiar people; as mentioned above, in absence of familiar people and with perception of low level of physical danger, people were less collaborative and less likely to take part in rescue operations. The results of the analysis were combined in a new four-fold typology model that describes the individual and collective reactions focused on solidarity (Fig. 4).
Respondents who were in presence of attachment figures and with perception of mild physical danger display attachment behavior. Individuals seemed to seek and reinforced the pre-existing group by relying upon a leader and leaving the scene when the members were safe; more, they behaved actively taking part in relief activities but collaborated less with strangers.
The perception of mild physical danger combined with the absence of attachment figures nearby, especially a leader, made the milling process and the reduction of uncertainty longer: this pushed people to collect further information. However, there were few possibilities to engage in response behavior, interacting with strangers and taking part in rescue activities.
Respondents in presence of familiar persons and with a perception of high physical danger decided to stay in the aftermath and reacted very actively. Again, the figure of the leader assumed an important role. They strongly committed to in relief activities, even helping and collaborating with strangers and unknown people.
People that perceived physical danger as high and with no familiar people in the proximity were collaborative and likely to interact with people around, even though they were strangers. Possibly, a strong interaction with the other victims was the result of an increasing sharing of feeling and common faith.
This study confirmed that people express strong solidarity, helping both attachment figures and also unknown people; it also underlined the affiliative and social nature of the spontaneous reaction. Nevertheless, survivors’ behaviors in the immediate aftermath of a disaster are strictly related to the presence or absence of attachment figures; when victims are with relatives or familiar people, they are more likely to behave within the group, relying on a leader. They decide to react actively but they are focused on helping and collaborating mostly with familiar people. People with strangers or alone instead tend to be more proactive and altruistic, helping people around. Results also underlined the importance of the environmental assessment in solidarity; the reaction of people is also influenced by the surrounding situation and the higher was the perception of physical danger, the more intense was the reaction and the solidarity showed.
This study also presents some limitations; this research was conducted in the immediate aftermath of the disaster; therefore, the chaotic situation made the data collection very challenging. However, the sample was not meant to be representative of the overall population as the non-probabilistic sampling methodology demonstrates; on the contrary, this work might serve as a pilot study for more systematic and scientific research in different cultural context and during different type of events.
This paper should not be considered as an attempt to develop a new theory. On the contrary, it is an attempt to corroborate the already existing literature with empirical data that lack; affiliation approach and group approach can be useful to describe survivors’ behavior in the immediate aftermath of a disaster and it should be improved with specific disaster related factors, such as the danger perception or the environmental assessment. Furthermore, it might serve as a pilot study in advance of a more systematic and scientific research in different cultural context and during different type of events. Investigating how people react in the immediate aftermath of disasters, in fact, is essential to develop a more efficient and effective response strategy by disaster planners. The very first phase after the event is the most concerning one because usually authorities have little control over the situation due to complexity and uncertainty and actions or inactions, often uncontrolled and unmonitored, affect subsequent relief operations. Usually relief teams need time to arrive to the scene and people rely on themselves or on others in the proximity. The natural tendency to help others can be essential to reduce losses and to fill the temporal gap between the event and the arrival of the organized relief unit. This spontaneous behavior was characterized by collaboration and help: this is important in order to consider survivors not as a concern but as a resource.
Dr. Andrea Bartolucci: firstname.lastname@example.org
The data used can be accessed via https://dx.doi.org/10.6084/m9.figshare.4047561.v1
The authors have declared that no competing interests exist.
AcknowledgementsThis project was carried out during the European Relief Mission, coordinated by EU Mechanism, in the aftermath of the Typhoon Haiyan (Yolanda) that hit the Philippines on November 2013. We would like to express our gratitude to the EU Mechanism, The Italian Civil Protection Department, the Marche Region Civil Protection Department and the Ares Marche Association. A heartily gratitude goes to Ms. Ofelia Ortega, the Municipal Local Government Operation Department and to all the Tanauan authorities who gave us the opportunity to administer the questionnaire to the affected population, organizing meeting and interviews. Last but not least, a special thanks goes to all the questionnaire respondents without whom this project will not be possible.
- Aguirre, B. E., Wenger, D., Glass, T. A., Diaz-Murillo, M., & Vigo, G. (1995). The social organization of search and rescue: evidence from the Guadalajara gasoline explosion. International Journal of Mass Emergencies and Disasters, 13(1), 67–92.
- Quarantelli, E. (2001). THE SOCIOLOGY OF PANIC.
- Clarke, L. (2002). Panic: myth or reality? Contexts, 21–26.
- Quarantelli, E. L. (1989). How Individuals And Groups React During Disasters: Planning And Implications For EMS Delivery.
- Quarantelli, E. (2001). THE SOCIOLOGY OF PANIC.
- Auf der Heide, E. (2004). Common misconceptions about disasters: Panic, the “disaster syndrome,” and looting. The First 72 Hours, 340–380.
- Johnson, N. R. (1987). Panic and the Breakdown of Social Order: Popular Myth, Social Theory, Empirical Evidence. Sociological Focus, 20(3), 171–183.
- Quarantelli, E. L. (2008). Conventional Beliefs and Counterintuitive Realities. Social Research, 75(3), 873–905.
- Glass, T. a. (2001). Understanding public response to disasters. Public Health Reports (Washington, D.C. : 1974), 116 Suppl , 69–73.
- Helsloot, I., & Ruitenberg, A. (2004). Citizen Response to Disasters: a Survey of Literature and Some Practical Implications. Journal of Contingencies and Crisis Management, 12(3), 98–111.
- Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Basil Blackwell. Chicago
- Perry, R. W. (1979). Evacuation decision-making in natural disasters. Mass Emergencies, 4, 25–38.
- Riad, J. K., Norris, F. H., & Ruback, R. B. (1999). Predicting Evacuation in Two Major Disasters: Risk Perception, Social Influence, and Access to Resources1. Journal of Applied Social Psychology, 29(5), 918–934.
- Turner, J. C., & Oakes, P. J. (1986). The significance of the social identity concept for social psychology with reference to individualism, interactionism and social influence. British Journal of Social Psychology, 25(3), 237–252.
- Aveni, A. (1977). The not-so-lonely crowd: Friendship groups in collective behavior. Sociometry, 40(1), 96–99
- Drury, J., Cocking, C., & Reicher, S. (2009). The Nature of Collective Resilience: Survivor Reactions to the 2005 London Bombings. International Journal of Mass Emergencies and Disasters, 27(1), 66–95.
- Alexander, D. E. (2013). Paradoxes and Perceptions - Four essays on disaster. IRDR Occasional Paper 2013-01, 1–55.
- Durodié, B., & Wessely, S. (2002). Resilience or panic? The public and terrorist attack. Lancet, 360(9349), 1901–2.
- Lowe, S., & Fothergill, A. (2003). A need to help: Emergent volunteer behavior after September 11th. Beyond September 11th: An account of post-disaster research, 293-314.
- Mawson, A. R. (2005). Understanding mass panic and other collective responses to threat and disaster. Psychiatry, 68(2), 95–113.
- Aguirre, B. E. (2005). Emergency evacuations, panic, and social psychology. Psychiatry, 68(2), 121-129.
- Dynes, R. (2008). Mass Panic and Social Attachment: The Dynamics of Human Behavior by Anthony R. Mawson. American Sociological Association, 37(2), 138–139.
- Chmiliar, L. (2010). Case study surveys. In A. Mills, G. Durepos, & E. Wiebe (Eds.), Encyclopedia of case study research. (pp. 125-127). Thousand Oaks, CA: SAGE Publications
- National Disaster Risk Reduction and Management Council of the Philipines. (2014). Updates re the Effects of Typhon “Yolanda” (Haiyan).
- Bourque, L. B., Shoaf, K. I., & Nguyen, L. H. (1997). Survey research. International Journal of Mass Emergencies and Disasters, 15(1), 71–101.
- Teddlie, C., & Yu, F. (2007). Mixed Methods Sampling: A Typology With Examples. Journal of Mixed Methods Research, 1(1), 77–100
- Roberts, C. (2007). Mixing modes of data collection in surveys: A methodological review. Discussion Paper. N/A. (Unpublished)
- NSO Philippines. (2000). Philippine Census, 2000. Household Population by Ethnicity, Sex and Region.