Research in Emergency Aftershock Response (REAR) Literature Review: Disaster Response, Gender, Resilience, and Appropriate Technology Use

I have been working the last six months or so on a NERC funded project at the Centre for Research in Digital Education at the University of Edinburgh titled REAR (Research in Emergency Aftershock Response). It is an interdisciplinary (geosciences, data science. infomatics, cultural influences, digital education, humanities) and inter-organisational (Edinburgh, Leeds Beckett, Ulster, Plymouth universities along with the British Geological Survey and Concern Worldwide) project. It is designed to explore:

Correct community response to aftershocks reduces subsequent loss of life in earthquake-stricken areas but local response can be delayed or inhibited by social, cultural and political factors. Effective, large-scale user engagement with appropriate information, so essential in emergency response, requires work on developing public awareness at scale, designing effective co-learning across multiple stakeholder groups, and building a deep understanding of the social and gender issues which might limit, or enable, user engagement.

Aftershocks are generally misunderstood (even by me). My fieldwork in Nepal as part of this project served to illustrate that. In a country still recovering from the earthquakes of 2015, we experienced aftershocks in February of 2017. Two years and hundreds of aftershocks later, the effects were still being felt.

The role of digital education in this process is to determine current technological use towards education and/or communication in crisis, cultural and/or sociological constraints in that process (ie, gender differentials in ICT use and reasons behind that), and the hybrid (digital and face to face) learning systems that might be developed or encouraged as a result. ICT use in specific locales is complicated enough from a research perspective; now consider what communication and information authority look like in the wake of disasters. Processes break down and are reassembled on the spot; misinformation is rampant; the gaps in the social nets become chasms; women and children are routinely left behind. It is a complicated environment to research, let alone augment local capacity for disaster response.

Also, these interdisciplinary projects directed towards development or emergency response needs are a welcome step. In my experience outside of academia, many of these projects flounder precisely due to the lack of a cultural and educational component to align them with local practice. So there are disconnects between the well-meaning, authoritative, and rigorous scientific research and the capacity of local populations to use that research towards resilience. Much of this is aligned with larger research and funding patterns precipitated by the Global Challenges Research Fund (GCRF), a 5-year £1.5Bn fund that aims to ensure that UK research takes a leading role in addressing the problems faced by developing countries through “challenge-led disciplinary and interdisciplinary research; and strengthening capacity for research and innovation within both the UK and developing countries.” These interdisciplinary approaches to development challenges are or will be the norm so the project itself is a productive exploration as to how that could be arranged across sectors, organisations, and disciplines. It takes a fair amount of wrangling but ultimately produces a much more nuanced approach to disaster response.

We are in the process of working on the outputs for the project, but the literature review my team performed (the Engagement and Learning team, one of four teams) might be of interest to some. We explored a variety of approaches to ICT-based responses to disaster response and more and more has been released in the interim since drafting this (the Handbook on Data Protection in Humanitarian Action springs to mind in response to the relative paucity of ethical discussion in the DRR literature). The review is below and there is a bibliography at the end for any and all interested. For brevity’s sake, the main findings are:

• Few rigorous experimental studies or quasi-experimental studies conducted in countries affected by earthquakes assess the effects of interventions on community response for women and children.
• Lack of generalisability: when evidence is robust, the attending study tends towards a high level of granularity.
• A reliance on resilience as an explicit objective: aside from some limited research, a general lack of criticality towards resilience as the unifying outcome
• Innovation with ICT by some, but ICT can increase divide if not linked to local conditions; tech drops are highly problematic in these disaster contexts
• What technology provides: “while responders tend to see communication as a process either of delivering information (‘messaging’) or extracting it, disaster survivors seem to see the ability to communicate and the process of communication itself as every bit as important as the information delivered.” Communication needs to be clear and concise. Science doesn’t like to present probability models (likelihood of disasters occurring again)
• Role of social media: problematic but ultimately necessary. There are local and often gender specific approaches to using social media in specific contexts. Women less likely in certain locations to use broadcast social media (Twitter, Facebook), opting for more intimate, generally secure messaging applications (Whatsapp, etc.).
• Gender based access is problematic depending on locale; it must be designed on and data must be disaggregated. Children are the same; they cannot be some “Other” category. Women in disaster situations often drop off the radar precisely because the data isn’t being disaggregated according to gender.

Introduction
Large earthquakes (and secondary hazards such as aftershocks, tsunamis and earthquake-triggered landslides) continue to cause some of the most devastating humanitarian disasters, yet they remain poorly understood by humanitarian organisations. This is partly because of the unique features of major earthquakes compared to other disasters (International Federation of Red Cross and Red Crescent Societies, 2012). They are unpredictable, low-frequency, high-impact events with little or no warning and very rapid onset. Some level of preparedness and disaster risk reduction is possible in at-risk areas (risk mapping and targeted mitigation, awareness-building, modification of building codes and logistical response capacities, for example) (Concern Worldwide, 2015 ), but decision-making in the response phase immediately after a major earthquake has often been reactive and ad hoc, rather than evidence-based and strategic. As such, the purpose of this literature review is to evaluate research that parallels such efforts as described in collaboration with Concern Worldwide, to identify gaps in the literature, and to draw methodologies from the research that have proven effective in similar contexts. To do this, this literature review is organised in thematic sections detailing educational initiatives, gender issues, and ICT-based responses.

Methods
This literature review has identified hundreds of academic articles, project sites, reports, and white papers on the topic of community responses to earthquakes through desk research searching academic databases with keyword searches, through online searches of relevant agencies and projects, through discussions and interviews with those working in ICT and development fields (NGOs, government agencies, and private firms or organisations). Search methods included keyword and author searches, citation searches, and reverse citation searches. For non-academic sources, search methods included an array of organisations that would nominally work in or suppose those working in disaster response scenarios either directly as disaster response or development professionals or indirectly as educational, health, or related organisations. These include national and international organisations such as UNESCO, DFID, USAID, USGIS, GIZ, and more, as well as educational and media-based organisations like British Council, BBC Media Action, the CDAC Network, and more.

This literature review prioritises research that employs community responses to earthquakes that have within them some measure of technology use and those projects that specifically target often marginalised communities (depending on the region), such as women and children, communities that are often most adversely affected in the event of a crisis. We have given weight to peer-reviewed publications, but not exclusively so as we recognise that many of the more innovative approaches to disaster response have yet to be evaluated through academic research, but have often been evaluated through monitoring & evaluation practices consistent with the practices of NGOs and INGOs, where third party evaluation is often explicit. The use of ICT for disaster response has a considerable research body of literature that is being drawn on to inform this literature review as well as further development of the REAR project itself.

Educational Interventions
Transitioning into research on educational initiatives requires a brief note on the scope of what education itself implies as presented in this literature review. As defined here, educational intervention refers to a specific program or set of steps to help a region, community, or individuals improve in an area of need. Even within the context of aftershock response, these areas of need are numerous. There are stated areas of need for disaster response reduction (DRR), whose aims are to reduce the damage caused by natural hazards like earthquakes, floods, droughts and cyclones, through an ethic of prevention (UNISDR). The educational initiatives designed to address this DRR need are designed, according to UNESCO (n.d.), to save lives and prevent injuries should a hazardous event occur; prevent interruptions to the provision of education, or ensure its swift resumption in the event of an interruption; and develop a resilient population that is able to reduce the economic, social and cultural impacts should a hazardous event occur. Further, education of this sort takes into account “the relationships between society, environment, economy, and culture and their impacts”, while promoting “critical thinking and problem-solving as well as social and emotional life skills that are essential to the empowerment of groups threatened or affected by disasters” (ibid). This position emphasizes immediate safety, continuity, and resiliency.

Much of this position is encapsulated in The Sendai Framework is a 15-year, voluntary, non-binding agreement which recognizes that the State’s aims for the “the substantial reduction of disaster risk and losses in lives, livelihoods and health and in the economic, physical, social, cultural and environmental assets of persons, businesses, communities and countries.” Targets that speak to this reduction of disaster risk and losses include local indicators such as disaster mortality, number of affected people, economic loss, damage to critical infrastructure, disruption of basic services. At the national level, indicators include the availability of a disaster risk reduction strategies, international cooperation, and availability of early warning systems and assessments (UNISDR, 2015). The Sendai Framework prioritises disaster risk awareness, risk governance, resilience investments, disaster preparedness, and reconstruction.

Throughout these frameworks is the emphasis placed on intersectionality; that there must be “integrated, multi-sectoral approaches to DRR, and to strengthening DRR in key sectors, such as education, agriculture and health” and that “development and resilience are unlikely to be sustained unless disaster risk is explicitly addressed in all development initiatives” (GPDRR, 2013). Education, as such, is one strand of a larger response to disaster. Such a position, while holistic, requires further nuance as applied to aftershock response. Education can be designed for DRR practitioners: international and domestic NGOs, government agencies, and humanitarian organisations and their focus on governance, preparedness, resilience investments, and reconstruction. It can be designed for local communities and the community organisations that serve them: schools, hospitals, transportation and infrastructure authorities, and further community organisations and their focus on awareness, building local resilience, and to a lesser extend reconstruction. All require education contextualized to need and capacity.

Camacho et al (2016) note the need for frameworks that operationalise this need for intersectionality. Educational practices in these competency-based models are largely understudied, although familiar methods such as simulations, peer-driven models, and broad stakeholder models of education are presumably commonplace. Camacho et al (2016) suggest ensuring professional competence by supporting adaptation of technical and non-technical professional capacities into low-resource and emergency contexts; and preparing for effective team performance in the field. Phibbs et al (2015) caution for inclusion of disabled people in emergency educational initiatives noting the lack of impact within this community in response to the Christchurch earthquakes of 2010; further educational initiatives demonstrated a correlation between these and disaster impact, ultimately suggesting that the impact of disasters can be substantially reduced if people are well informed and motivated towards resilience. The authors suggest embedding such educational initiatives in school curricula, as well as within the programs of government and non-government agencies.

Shaw, Kobayashi & Kobayashi (2004) suggest a parallel tack, that of embedding education in family and community education structures, in concert with school curricula or with government or non-government agencies. The authors suggest that a culture of disaster preparedness is possible with a series of educational methods, from discussion, experiential learning, and visual aids to family and community education initiatives. Based on survey results from students in earthquake zones of Japan, the findings suggested that lived earthquake experience does not inherently lead to greater awareness and resilience, but rather family and community education initiatives. School education was found to be most effective when based on conversational methods, and media-based programming, particularly television. Hachiya & Akashi (2016) suggest the need for a diversity of educational initiatives aimed at different stakeholders in the disaster response process would prove most effective; the authors suggest stakeholders relevant to the disaster being responded to, namely the Fukushima disaster of 2011. These groups include for experts/professionals, teachers, students, risk communicators and others. While these groups would vary depending on locale and disaster being responded to, ultimately there is the explicit need presented for educational differentiation.

Masuzawa et al (2014) suggest an experiential training and education programme towards what they call response literacy, but most of the literature refers to as resilience or emergency preparedness, provides an effective and accessible approach to disasters. However, the impact of these educational initiatives are often gleaned from self-reported survey instruments, one that does little to disaggregate the cultural elements of disaster response thereby limiting its generalizability. Yet, the research is suggesting that experiential and multi-stakeholder educational initiatives can prove effective in stimulating effective disaster response. Alexander (2003) argues the need, echoed in much of the research subsequently, for more uniform standards in disaster education, while Nathe (2000) presents the need for tailored communication in disaster education, one that reverberates throughout the literature and is revisited in subsequent sections of this literature review. Nathe provides recommendations that anchor much of the work being undertaken as part of the REAR project; for education to be effective, it must be accessible and actionable, and ultimately urgent.

According to Nathe, communication must be highly accessible:

“Simple language in manageable amounts is absolutely necessary. Though credentialed spokespersons are one of the most important sources of information, specialists who speak only in the jargon of their discipline will not be effective. Authoritative interpreters of technical information should be cultivated, encouraged, and paid well. Fit the specialist to the topic: geologists and seismologists should talk about earth science, engineers and architects should talk about structures, and firefighters and emergency responders should talk about home safety and neighborhood organization.” (p. 194).

Probability models, however contentious, should not be avoided as they represent conduits for behavioral change: “Probability estimates will not, in themselves, motivate people to take action, but the information will assist in creating the uncertainty that is so important to behavior change. Earthquake prediction is a very inexact science, but where geoscientists have some understanding of the behavior of specific faults and the frequency of quakes on them, they should offer these rough forecasts.” Broader educational campaigns and initiatives must be “coherent and collaborative, their information must be credible and understandable, and the information must reach its intended audience. In that statement is a prescription for close cooperation among technical specialists and educators, constant communication among educational organizations, and sophistication and creativity in the message translators and communicators” (p. 195). Nathe suggests this is altogether a complicated process, particularly in locales where authority is diffused through an array of community actors and cultural factors, but ultimately a necessary one.

Gender Issues
Women and children often suffer a disproportionate number of casualties in earthquakes and aftershocks, often due to their marginalized status within affected communities. This suggests, a suggestion borne out in the literature, the need for differentiation in identifying methods for effective community response, differentiation that has proven elusive in many disaster response efforts. “Dealing with difference represents a significant challenge for disaster managers, one that remains largely unrecognized or suppressed in favor of a sometimes spurious egalitarianism which attempts to treat everyone the same” (Fordham, 1999). However, women and children both have experiences “embedded in socially constructed modes of living” often rendering them “chronically vulnerable”, a vulnerability exacerbated in times of disaster. These gender issues are problematised by a neglect of private space as an appropriate area of research, a neglect that proves particularly problematic in disaster response, and one that lends itself to chronic vulnerability precisely as it establishes an almost chronic invisibility for many women (Fordham, 1998).

This chronic vulnerability can be exacerbated by ICT-based approaches to disaster response precisely due to the gender digital divide that exists in much of the world. Overall, women are 14% less likely to own a mobile phone than a man, a percentage that equates to 200 million fewer women than men owning mobiles. This gender digital divide is particularly acute in South Asia where 72% of the female population is unconnected (Santosham & Lindsey, 2015). Yet, ICT ownership is not the only issue contributing to this chronic vulnerability. Literacy rates globally tend to disadvantage women, rendering SMS or text-based communication problematic. Technical literacy issues, e.g. the capacity to send and receive SMS, further makes ICT-based disaster response disproportionately beneficial to men. There is a strong positive correlation between areas where educational gender gaps are highest and regions where the gender digital divide is greatest (Web Foundation, 2015).

Further complicating ICT based responses are gender-based divides in terms of use preferences. According to Santosham & Lindsey (2015), women use mobile services, such as SMS and mobile internet, apart from voice less frequently and intensively than men. This is, at least partly, due to the entanglement of cultural and economic issues: the cost of data, for example, and the tendency of women in many parts of the world to only buy from female SIM and data card vendors leads to a context where women are often less likely to use the phone. Yet, it is clear from the research that the demand for ICT use and greater social participation is present, if muted by sociocultural and economic contexts. This demand, the research suggests, can be activated through digital literacy training, public private partnerships, locally relevant and content (Alliance for Affordable Internet, 2014), as well “greater investment in girls’ education to work hand-in-hand with targeted ICT skills programmes” (Web Foundation, 2015).

Women are much less likely to use particular social media, such as Facebook, than men, particularly in South Asia; a recent study in India had 85% of men mostly using Facebook compare to 47% of women (Schoemaker, 2015). Women were much more likely to use Whatsapp than men in South Asia, a trend that emerges from social and cultural contexts and one with great relevance to disaster response in particular regions. Social media use, and indeed most technology use, supports and often reinforces existing cultural practice. Facebook, in its emphasis on open communication with a broader network some of which might lack familiarity, runs counter to existing cultural practice for women in many parts of the world. Hence, Whatsapp is more popular for women with its emphasis on intimate networks and communication. This is problematic insofar as much social media use in disaster response projects has involved social media with a broadcast component: Twitter data being publicly available, for example, forms a great deal of the research literature in its use in disaster response. Female users rely heavily on social networks to learn – but barriers of social norms in particular regions around males fears of women using internet and the lack of skills amongst their helpers inhibit that learning. In general, it is very uncommon for women to seek help from mobile phone retailers or community resource people (GSMA Connected Women, 2015).

Much of the chronic vulnerability experienced by women extends beyond ICT. There is a need for gender disaggregated data across the spectrum of humanitarian response and development, a need reiterated by the Alliance for Affordable Internet (2014), without which gaps in access to ICTs cannot be addressed. This need for gender disaggregated data is often a supply issue: many organisations generating data, particularly in disaster response situations, do not collect it (Bradshaw & Fordham, 2015). This lack of gender disaggregated data creates gaps in information that limits the effectiveness of humanitarian response in the event of a disaster (Mazurana, Benelli & Walker, 2013), rendering women more chronically vulnerable precisely due to their statistical invisibility. ICT-based responses to disasters relying on data that is not gender disaggregated will tend to favor men due to a range of economic, social, and cultural factors. The call for gender disaggregated data has been reiterated in the research (Christie, 2016) as well as in the outputs of large national level development organisations, such as USAID’s Data Disaggregation by Geographic Location (2016) and and DFID’s Strategic Vision for Women and Girls (2014). Ultimately, the data provides indications of gendered differences in disaster effects, such as the disproportionate mortality rates discussed in Neumayer & Plumper (2007). This was borne out in Nepal context in response to the 2015 earthquake, where higher female mortality rate during the earthquakes can be attributed to numerous factors such as “male migration to the capital city and abroad and traditional gender roles whereby women and girls were more likely to be inside the house than men. In many cases, women were reported to have delayed their escape to rescue their children, older family members and valuables” (Limbu, 2016). Without gender disaggregated data, crafting appropriate disaster responses would fail to account for these accounts of self-sacrifice and gender-based perceptions of risk.

The analysis that gender disaggregated data might be able to provide may in turn lead to actionable programmatic shifts in disaster response, such as a much greater focus and attention during risk assessments on assessing the specific vulnerabilities faced by women and girls and the extent to which these are being addressed by governments, donors and agencies; as well as the need for separating women and children in the data (Bradshaw & Fordham, 2013). It is clear from this research that any disaster response with or without the use of ICT will require data that is gender disaggregated as well as program activity that is responsive to risk perceptions and ICT use.

ICT & Media-based responses
The use of information and communications technology (ICT) in disaster response has a considerable body of research associated with it. While much of the research superficially presents ICT as a panacea for all disaster response, neglecting many of the social or cultural issues that govern its use, some of the research has moved beyond this simplistic treatment with criticality. What is clear in the literature is that the use of ICT in disaster response is accelerating. “ICT use for humanitarian response runs the gamut from satellite imagery to drone deployment; to tablet and smartphone use; to crowd mapping and aggregation of big data. Humanitarian actors applying these technologies include front-line responders in NGOs and the UN but also, increasingly, volunteers and the private sector” (Raymond, Howarth, & Hutson 2012). While the use of technology in disaster response has rapidly accelerated along with the number of actors employing this technology, the ethical and technical guidance available to practitioners has lagged considerably behind. While this literature review attempts to chronicle this technology use, it is important to note that there is a ‘doctrine gap’ between the adoption of technology in disaster response and the development of standards to govern its use (Raymond & Card, 2015). While the question of ethical use of ICT is addressed further in this literature review, the literature reviewed here reflects this lack of doctrine: many of the uses of ICT in disaster response lack either generalizability beyond the locale of their implementation, and lack sustainability beyond their initial project lifespan. As such, transformative change in the local community’s capacity for developing resilience in the wake of disasters is often negligible.

A further caveat to the use of ICT in disaster response is the general predilection towards what Evgeny Morozov calls ‘solutionism’, “which he describes as recognising ‘problems as problems based on just one criterion: whether they are “solvable” with a nice and clean technological solution at our disposal’” (2015). As the REAR project is attempting to engage meaningfully in disaster response, an issue that evades categorization as being fully “solvable”, this literature review will be keen to identify ICT based solutions that are subject to this solvability conceit as well as those that look to address disaster response issues beyond mere ICT reductionism.

The use of ICT in disaster response has a long and rapidly accelerating tradition; however, the monitoring & evaluation (M&E) of that ICT use on the impact of disaster response is a relatively new phenomena. As such, there is in the literature a large body of work dedicated to ICT uses in disaster response without, barring some exceptions, the level of scrutiny needed to determine whether that ICT use was indeed beneficial. The research discussed in this section has been selected precisely because it offers an evaluation framework of its own ICT use. Haataja, Hyvärinen & Laajalahti (2014) presents a study exploring the communication patterns of non-DRR practitioners in crisis situations, their communication needs, and how ICT meets those needs. The aim of this research mirrored much of what the REAR project is attempting to do: “to gain further insight into what solutions could match the existing communication needs of individuals and how the acceptance of relevant upcoming technology could be enhanced among potential future adopters.” Although drawing on a relatively meagre data sample emerging solely from three focus groups, findings indicated the need for timely and relevant information in response to disasters that spans, among other topics, “the magnitude and duration of the incident, precautionary actions that could be taken, and details on where to find additional information on the subject.” ICT use generally involved a mobile phone and tended towards a desire to be able communicate directly with a human representative from a responsible organisation for reliable information. Further was the desire to communicate with neighbors, friends, and family to disseminate or collect further information directly via mobile phone, rather than through a reliance on social media. Radio was also presented as a viable means for presenting information. However, the most telling finding and indeed the most relevant for the authority of the data itself was the tendencies of individuals to seek information from multiple sources to develop a comprehensive understanding of the disaster, thus suggesting the need for careful orchestration from actors in a larger response network. Relevant suggestions include more specific targeting of notifications and integrating notifications with other sources of information that are used already (e.g., weather and route planning applications).

Vos & Sullivan (2014) return to the theme of resilience as an organising construct of disaster response but draws reference to indigenous ICT solutions that have arisen in direct response to natural or manmade disasters including, the Sahana open source platform, which emerged from Sri Lanka in the aftermath of the 2004 Indian Ocean tsunami and has been used in multiple international disasters, and Ushahidi, also an open source platform, which was developed in response to the 2008 post-election violence in Kenya. Both represent a push towards co-designed ICT in response to disasters and both employ mobile phones and text messaging. Linnell (2014), based on the ICT based practices of disaster responders, suggests categories of participation for potential voluntary engagement: organised volunteers, semi-organised individuals, and non-organised individuals. These categories provide utility in differentiating between potential stakeholders in disaster response initiatives and their subsequent needs, but also suggest the varying degrees of systematic effort necessary to engage them in any aftershock-based response. Palen et al (2010) reiterate this need for differentiation in target groups, a differentiation that can only be realistically understood through a careful analysis of the information sharing activities of the larger community through ICT. Palen et al (2010) also highlight the difficulties in community driven ICT based responses to disasters, noting their susceptibility towards a lack of authority due precisely to their openness: for example, users have the capacity to undermine authority injecting false and misleading information into information streams, maliciously or not. This tendency is supported by much of the literature regarding the use of social media in disaster response: its impact is dependent, among other factors, on the perceived authority of the information being circulated.

However, despite these limitations, the need is reiterated throughout the literature for contextual adaptations to response, ie that responses emerge from and support appropriate responses to local contexts. The CDAC Network in their Media Landscape Guides for CDC, Ivory Coast, Indonesia, South Sudan, and Ethiopia, suggest as much, going so far as to state that “most of the creativity and novelty present in the response involved tactical adaptations to the context” and that “many ‘visible’ and high-profile innovations had no or little connection to the operational setting.” Beyond interjecting misleading information into disaster response contexts, many initiatives are showcases for technology, as “numerous examples were given of innovative ideas that were showcased in the media and in fairs in universities, but which had not been adopted by any operational organisation and had not seen any real-world testing.” As such, the literature suggests that innovative responses to local need are predicated on these “tactical adaptations” based on end-user testing in situ. These responses are in direct response to existing ICT practices based on thorough appraisals of ICT and informational need.

Reinforcing this call for tactical adaptations several efforts exist to visualise and dashboard data for disaster response practitioners and volunteers that pay some degree of fidelity to local community practice, or suggest the need for such fidelity. Herranz et al (2014) suggests that current technological advances on mobile and social computing could conceivably empower communities of volunteers and disaster response practitioners to participate more actively and to receive additional information from external sources, yet the authors are quick to note the intrinsic complexity of this task particularly as it applies to community membership, collaborative discussions, and awareness. Data contributed to mapping or visualisation efforts must come from multiple sources, including citizen-generated data as well as simultaneous pieces of information of different priority levels and at various geographic locations; and must present some credibility. The authors present in this research a design prototype for managing these elements towards a dashboard approach to both community contributed and scientific data towards disaster response. However, the authors are quick to note that such a prototype “cannot be tested in complex real situations where agencies and corps are not willing to try new technologies but to be as efficient as possible applying the procedures they are familiar with”, thereby limiting evaluation to controlled experiments and suggesting broader application is an ongoing, incremental, and community-driven engagement. Herranz, Gomez, & Diaz (2014) expand on the dashboard model further noting that empowering communities to use these tools and contribute data in response to disasters requires developing “representations that enable collaborative reflection, promote mutual visibility of volunteers’ efforts and sustain a shared view of the community.” The authors begin to wade through the complexity of community contributed information noting the varying levels of authority and reliability in varying streams of information: from vetted government or scientific agencies to social information contributed directly to social information gleaned from social media, all ascribe to varying structures of trust. This research provides the REAR project utility by clearly identifying the need to situate each stream of dashboard data within a reliability categorization. While beyond the scope of Herranz, Gomez, & Diaz’s research, further utility to REAR suggests the need for an algorithmic structure to the dashboard model with varying weights applied to varying streams of data depending on reliability measures, particularly if crowdsourced data will be included in the dashboard.

Xu, Nyerges, & Nie (2014) present a model for earthquake response in China, noting the importance before any technological intervention of emergency planning and emergency command systems, again reiterating a theme in the literature for understanding in situ conditions and communication practices and having these coded into planning and policy ahead of any technological intervention. The authors propose, again as a prototype, a conceptual model for earthquake disaster emergency response (EDER) along with basic modeling prototypes in response to a direct need as they see it: that earthquake data, particularly of the community contributed kind, is irregular texts or simple database schemes, resulting in difficulties in its use through technology and subsequently by disaster response practitioners and communities alike. Essentially, the research suggests that without formalised data, subsequent ICT interventions have minimal impact. However, the modeling outputs as discussed in the paper are prototypes and require validation in practice.

Yuan et al (2013) propose a similar model to community contributed data for disaster response in their human sensor network model, a means of capturing data from grassroots observers in the disaster areas. While suggesting the utility of such a model, the authors point to similar HNI applications such as Ushahidi, a software platform aggregating and mapping information provided by human sensors. Ushahidi pulls information from social media and SMS to create reports that can be placed on a web‑based, interactive map. Further applications include Tweak the Tweet (TtT), which relies on Twitter for data collection in the event of disasters. TtT relies on community use of a specific syntax for communicating during disasters based on based on primary or main hashtags that can be used to indicate the “who, what, and where” of the Twitter message content. For example, #name or #contact can be used to indicate “who”; #need, #shelter, #road, #open, #damaged can be used to indicate “what”; and #loc can used indicate “where”. These hashtags are used in conjunction with an event tag to organize the crisis, creating what is essentially a hybrid digital and manual disaster response system. The authors further point to OpenStreetMap and CrisisCommons, a platform for communication between HSN and disaster response practitioners. Yet the authors fail to critique the quality of the data being contributed by the community, or present a model for communication flow with such a diverse set of stakeholders as Herranz, Gomez, & Diaz (2014) provided.

Gilmour (2016) presents the Open Cities Project in Nepal, a project that worked with university students and community groups to collect structural data for 2,256 schools and 350 health facilities in the Kathmandu Valley, creating a base map by digitizing over 100,000 building footprints, mapping the road network, and collecting information on other major points of interest. The Open Cities team also conducted significant outreach to universities, technical communities, and government in order to expand the OSM community. Over 2,300 individuals participated in OSM trainings or presentations during the first year of the project. Gilmour discusses how in response to the Nepal earthquakes of 2015, volunteers contributed data related to damage throughout the country to the extent that resources were redirected to certain areas based on their contributions. However, while such a model proves capable of identifying damaged areas, particularly in terms of marshalling resources to affected areas, it does less in regards to providing actionable information to affected communities. In short, this is a model whereby crowdsourced information is mapped for use by disaster response practitioners rather than by the general public.

The potential and applied use of social media in disaster response is prevalent in the research, more than can be adequately reviewed in this literature review. Crooks et al (2013), however, build on the human sensor network model in their investigation of using Twitter as a distributed sensor system for identification and localization of the impact area of a disaster event, yet position that contributed data as merely complementary to more authoritative sources of data. Again, there is an emphasis on practitioner rather than community response. Takahashi, Tandoc, & Carmichael (2015) take a different tack, focusing on the communication patterns during Typhoon Haiyan in the Philippines, presenting a categorical effort to identify the nature of Twitter activity: time, geographic location, stakeholders, and engagement. Ultimately, the authors noted that Twitter use in disasters is essentially for dissemination of second-hand information (amplification), coordinating relief efforts (marshalling of resources), and to memorialize those who had passed, suggesting mixed use that would need to be accounted for in any community driven model.

Burks, Miller, & Zadeh (2014) show that counts of geo-tagged Tweets at different radii around a particular site are good predictors of shaking intensity when used in combination with other seismic data. In practice, this is important because additional data sources such as on-the-ground inspections by engineers, measurements from recording stations not connected to the internet, and online surveys may not be available before some decisions about relief aid need to be made. The cautions against using community contributed data on its own, without correlation to more authoritative data sources, or without some analysis as to its reliability or relevance, appears again and again in the research. However, the potential merits of such an approach is evident.

Beyond models for community content are the use of readily available technologies in disaster response, including radio and mobile. This is evident in low-resource contexts where available technology in the wake of a disaster is limited or seemingly antiquated, and where the disaster itself might sever all other forms of technology from the internet. Radio and simple mobile technology are expanding; “thousands of radio stations in sub-Saharan Africa where once there were only a handful” (Harding, 2015). Radio and mobile remain dominant in many developing regions. However, with this expansion comes parallel effects as the growth of media is coupled with a growing distrust of the media; “media as an institution is distrusted by 60% of countries” surveyed in a recent poll (Edelmen, 2015). In a similar study, Gallup concluded that, “in half the countries surveyed, less than a majority of residents expressed confidence” in their media (English, 2007). The use of radio and mobile technology, particularly as a tool for dissemination of authoritative data, presents complications.

The CDAC Network attempts to offset much of this distrust in their Media Landscape Guides, providing as they do profiles of the main radio stations, TV channels, newspapers and news websites; the most popular and trusted sources of news and information; media outlets that command significant national, regional, ethnic and religious audiences; media outlets which have strong influence on opinion formation in the ruling elite; contact details for each media outlet; and radio stations and TV channel listings in the country. According to Gormley (2014), the CDAC Network ultimately promotes coordination and dialogue between relevant stakeholders actors to ensure that they work together to best serve the needs of communities affected by crisis.

Communications systems are highly localised and rapidly evolving due to access to technology and that those with the most local expertise are local professionals, including local journalists, IT specialists and the private sector (Robinson & Wall, 2012). Radio, in particular, has proven highly relevant in disaster response and there are many examples of its use in disaster response. In Nepal, Milijuli Nepali (Together Nepal) is a radio programme providing people affected by the Nepal earthquake with practical, actionable information – on shelter, sanitation, food, water, health and employment – for survival and recovery. Milijuli Nepali was informed by continuous needs assessments and audience feedback, incorporating secondary data sources and primary research conducted by BBC Media Action through Facebook and mobile app surveys in the most affected districts; information was shared with humanitarian actors through online dashboards (Lifeline) to help them tailor their communication to meet communities’ needs (BBC Media Action, 2016). Extensive evaluation of this programme through a mixed methods approach throughout Nepal (survey of 4000 people across 25 districts, focus group discussions, in-depth interviews, and more) noted an impact on media’s awareness on role they could play in disaster response. Further the programme had developed networks where community response was fostered, and national and local radio stations started producing their own disaster response content. However, the authors noted that getting information verified by humanitarian and government sources was challenging due to their limited accessibility.

In Haiti in 2010, as in other emergencies, communications infrastructure was seriously damaged by the earthquake. Several radio stations collapsed; many lost the ability to pay staff because their advertising base collapsed (Wall & Chery, 2011). One of the remaining radio stations became a beacon for community led response. Radio 1’s internet connection had survived the earthquake and a well known broadcaster and social media enthusiast with a strong Facebook presence and a Twitter feed began broadcasting. The broadcaster set up an information system designed to consolidate the incoming requests, attempting to respond to them systematically and sharing the information collected with audiences. This was in addition to a team of producers who interviewed all those arriving at the station (an average of 50 people every hour). “Each day, the compiled list of people who were being sought was given to the courier who went out on his motorbike to search for them. Simultaneously, the team’s uploaded this information to Facebook, relayed it on Twitter and broadcast it on air. The whole system ran for several weeks without any support from the international aid effort” (Robinson & Wall, 2012).

In South Sudan, organisations used a localised USB radio to inform people at protection of civilian sites about planned programmes, enabling communities to provide more active feedback. Daily shows were recorded at select locations at the site. In Somalia, ‘Radio Ergo’, a locally run Somali-language station, streams on shortwave radio in order to reach people in remote areas who are not served by FM radio. The show is based on local reporting gathered by a network of correspondents with themes including health, education, protection, agriculture and livestock and gender (Humanitarian Policy Group at ODI, 2016). There are other examples from the research suggesting the need for hybrid models employing readily accessible technology and platforms, such as radio and social media. The role of mobile technology is implicit in much of this research (as the social media is contributed via mobile technology), but is more explicit in health fields, such as the research documented by MacPherson & Chamberlin (2013) outlining the various projects that have used mobile technology to great effect.

Further research with ICTs emphasizes the role of gaming and simulations to improve resilience to and awareness of disasters, perhaps best represented by Agrawal et al (2016)’s review of the use of augmented reality in disaster response. The authors note the limitations of such approaches primarily due to elevated costs and the considerable expertise needed for their maintenance, rendering them highly unwieldy in disaster responses in poorly resourced environments. Anton (2015) explores the use of OERs in building disaster response training provisions through gaming simulations in Romania, but the scalability and relevance of such an approach to broader disaster response efforts remains hypothetical. Aerial robotics and drones potentially provide utility in disaster response, but remain largely about data collection rather than disaster response. However, some smaller studies have demonstrated potential use; Schroeder & Meyer (2016) discuss the use of a drone in Nepal to capture data related to damaged infrastructure, followed by analysis by Nepali citizens and scientists, along with international technology professionals and aid workers, providing a level of detail that had been previously elusive. The authors note several efforts that have proven complementary to the Nepalese one: the International Organisation for Migration (IOM) use of aerial robots in Haiti to capture aerial imagery to assess disaster damage and displacement; Medair and Catholic Relief Services (CRS) use of aerial robots to collect imagery to inform their reconstruction and rebuilding efforts following Typhoon Haiyan in the Philippines, and so forth. However, the authors conclude that the humanitarian community requires a more convincing evidence base before aerial robots are broadly adopted. The authors also note the need for ethical practice and codes of conduct in response these aerial robots

Ethics and Efficiency of ICT Use in DRR
The ethics of ICT use for disaster response is generally absent in the literature, although there has been a recent call for the development of robust ethical frameworks to guide practice. There is an ethical gap in the use of technologies in humanitarian practice, what Raymond & Harrity (2016) refer to as a ‘doctrine gap’, particularly as regards the use of data generated by multiple types of ICTs. These gaps span thematic areas:

• Rights, privacy and consent. Particularly in disaster response environments, informed consent and the right to privacy are often negligible.
• Data sharing and retention: organisations lack clear guidance about when and with whom they can share what forms of data, and little rigor for deciding what data from what sources should be retained, for how long, and for what purposes. (the recent Handbook on Data Protection in Humanitarian Action addresses some of this).
• Protection of vulnerable populations: “when it comes to data, however, there is no shared understanding of how certain types of data can increase the risks faced by certain groups.” Great care must be taken by disaster response practitioners to ensure risk is assessed and participants are aware of this risk. (Humanitarian Working Group at ODI, 2016)

As Ganesh, Deutch, & Schulte (2016) attest, marginalised people have different needs for privacy and security online, and these needs need to be integrated into any disaster response solution. The authors point to the need for “collaborations across and within technology and activist movements and communities must recognise their different histories of engagement with politics, technology and the state.” Practitioners in this space acting without a clear understanding of this larger social, political, and technological environments are acting unethically.

Many technologies are currently being used often “without clearly established protocols, ethical standards and objectives for what actionable information is most critical in specific scenarios. Social media tracking inherently excludes those without access to such technology” (Humanitarian Working Group at ODI, 2016). Concise ethical frameworks linked to a comprehensive understanding of the larger social, political, and technological environment is needed to account for these gaps in the literature.

Bibliography
Agrawal, A., Acharya, G., Balasubramanian, K., Agrawal, N., & Chaturvedi, R. (2016). A Review on the use of Augmented Reality to Generate Safety Awareness and Enhance Emergency Response.
Alexander, D. (2003). Towards the development of standards in emergency management training and education. Disaster Prevention and Management: An International Journal, 12(2), 113-123.
Alliance for Affordable Internet (2014). Affordability Report.
Anton, M. (2015). The Impact of Open Educational Resources in Educating Citizens on Emergency Response.
BBC Media Action (2016). What role does lifeline preparedness play in enabling effective communication in a crisis? Case studies from Myanmar and Nepal.
Bradshaw, S., & Fordham, M. (2015). Double disaster: disaster through a gender lens. Hazards, Risks and Disasters in Society, 233-251.
Bradshaw, S. & Fordham, M. (2013). Women, Girls, and Disasters: A review for DFID.
Burks, L., Miller, M., & Zadeh, R. (2014, July). Rapid estimate of ground shaking intensity by combining simple earthquake characteristics with tweets. In 10th US Nat. Conf. Earthquake Eng., Front. Earthquake Eng., Anchorage, AK, USA, Jul. 21Y25.
Camacho, N. A., Hughes, A., Burkle Jr, F. M., Ingrassia, P. L., Ragazzoni, L., Redmond, A., … & von Schreeb, J. (2016). Education and Training of Emergency Medical Teams: Recommendations for a Global Operational Learning Framework. PLoS Currents, 8.
Christie, M. E. (2016). Gender equity: Integration, coordination, resources, and support for gender research.
Comfort, L., Wisner, B., Cutter, S., Pulwarty, R., Hewitt, K., Oliver-Smith, A., & Krimgold, F. (1999). Reframing disaster policy: the global evolution of vulnerable communities. Environmental hazards, 1(1), 39-44.
Crooks, A., Croitoru, A., Stefanidis, A., & Radzikowski, J. (2013). # Earthquake: Twitter as a distributed sensor system. Transactions in GIS, 17(1), 124-147.
De Albuquerque, J. P., Herfort, B., Brenning, A., & Zipf, A. (2015). A geographic approach for combining social media and authoritative data towards identifying useful information for disaster management. International Journal of Geographical Information Science, 29(4), 667-689.
DFID (2014). Stepping-up a gear for girls and women: Updates to DFID’s Strategic Vision for Girls and Women, to 2020 and beyond.
English, C. (2007) Quality and Integrity of World’s Media Questioned [online]. Gallup, 17 December. Available: http://www.gallup.com/poll/103300/quality-integrityworlds-media-questioned.aspx.
Fordham, M. (1999). Balancing Resilience and Vulnerability. Int. J. Mass Emerg. Disasters, 17, 15-36.
Fordham, M. H. (1998). Making women visible in disasters: problematising the private domain. Disasters, 22(2), 126-143.
Ganesh, M. I., Deutch, J., & Schulte, J. (2016). Privacy, anonymity, visibility: dilemmas in tech use by marginalised communities.
Gilmour, E. (2016). Mapping for resilience: crowd-sourced mapping in crises. Humanitarian Practice Network.
Gormley, B. (2014). The changing face of humanitarian reporting–putting the survivor centre stage. The Future of Humanitarian Reporting. UK: City University, 81-85.
GPDRR (Global Platform on Disaster Risk Reduction) (2013). Chair’s summary. Fourth session of the global platform for disaster risk reduction. Geneva, 21–23 May 2013.
GSMA Connected Women (2015) Accelerating Digital Literacy: empowering women to use the mobile Internet.
Haataja, M., Hyvärinen, J., & Laajalahti, A. (2014). Citizens’ Communication Habits and Use of ICTs During Crises and Emergencies. Human Technology, 10(2).
Hachiya, M., & Akashi, M. (2016). Lessons Learned from the Accident at the Fukushima Dai-ichi Nuclear Power Plant—More than Basic Knowledge: Education and its Effects Improve the Preparedness and Response to Radiation Emergency. Radiation Protection Dosimetry, 171(1), 27-31.
Harding, P. (2015). Public service media in divided societies: Relic or renaissance. BBC Media Action.
Herranz, S., Romero-Gómez, R., Díaz, P., & Onorati, T. (2014). Multi-view visualizations for emergency communities of volunteers. Journal of Visual Languages & Computing, 25(6), 981-994.
Herranz, S., Gómez, R. R., & Díaz, P. (2014). Visualization Techniques to Empower Communities of Volunteers in Emergency Management. In DMS (pp. 335-344).
Humanitarian Policy Group at ODI (2016). Humanitarian Innovation: Humanitarian Exchange.
Limbu, S.T. (2016). Women in disaster: Gendered vulnerabilities and intersecting identities in the wake of the Nepali earthquakes.
Linnell, M. (2014). Citizen Response in Crisis: Individual and Collective Efforts to Enhance Community Resilience. Human Technology, 10(2).
MacPherson, Y., & Chamberlain, S. (2013). Health on the Move: Can Mobile Phones Save Lives. Policy Briefing, 7.
Masuzawa, Y., Hisada, Y., Murakami, M., Shindo, J., Miyamura, M., Suwa, H., … & Nakajima, Y. (2014). Practice on an Education and Training Program to Development of Response Literacy to Earthquake Disaster in a Central Business District in Japan. Journal of Disaster Research, 9(2), 216-236.
Mazurana, D., Benelli, P., & Walker, P. (2013). How sex‐and age‐disaggregated data and gender and generational analyses can improve humanitarian response. Disasters, 37(1).
Nathe, S. K. (2000). Public education for earthquake hazards. Natural Hazards Review, 1(4), 191-196.
Neumayer, E., & Plümper, T. (2007). The gendered nature of natural disasters: The impact of catastrophic events on the gender gap in life expectancy, 1981–2002. Annals of the Association of American Geographers, 97(3), 551-566.
Palen, L., Anderson, K. M., Mark, G., Martin, J., Sicker, D., Palmer, M., & Grunwald, D. (2010, April). A vision for technology-mediated support for public participation & assistance in mass emergencies & disasters. In Proceedings of the 2010 ACM-BCS visions of computer science conference (p. 8). British Computer Society. Chicago.
Phibbs, S. R., Good, G., Severinsen, C., Woodbury, E., & Williamson, K. (2015, January). Emergency preparedness and perceptions of vulnerability among disabled people following the Christchurch earthquakes: Applying lessons learnt to the Hyogo Framework for Action. In Australasian Journal of Disaster and Trauma Studies, IRDR Conference.
Raymond, N. A. & Harrity, C. S. (April 2016). Addressing the ‘doctrine gap’: professionalising the use of Information Communication Technologies in humanitarian action.
Raymond, N.A. & Card, B.L. (July 2015). Applying Humanitarian Principles to Current Uses of Information Communication Technologies: Gaps in Doctrine and Challenges to Practice.
Raymond, N.A.; Howarth, C. & Hutson, J. (2012). Crisis Mapping Needs an Ethical Compass, Global Brief. Available: http://globalbrief.ca.
Robinson, L., & Wall, I. (2012). Still left in the dark? How people in emergencies use communication to survive–and how humanitarian agencies can help. Tech. rept. 6. BBC Media Action, London, UK.
Santosham, S., & Lindsey, D. (2015). Connected Women 2015: Bridging the gender gap: Mobile access and usage in low-and middle-income countries.
Schoemaker, E. (2015). Facebook Domestication. Tanqeed.
Schroeder, A. & Meier, P. (2016). Automation for the people: opportunities and challenges of humanitarian robotics. Available: http://reliefweb.int/report/world/automation-people-opportunities-and-challenges-humanitarian-robotics
Shaw, R., Shiwaku Hirohide Kobayashi, K., & Kobayashi, M. (2004). Linking experience, education, perception and earthquake preparedness. Disaster Prevention and Management: An International Journal, 13(1), 39-49.
Takahashi, B., Tandoc, E. C., & Carmichael, C. (2015). Communicating on Twitter during a disaster: An analysis of tweets during Typhoon Haiyan in the Philippines. Computers in Human Behavior, 50, 392-398.
UNESCO (n.d.). Disaster Risk Reduction.
UNISDR (n.d.). What is Disaster Risk Reduction?
UNISDR (2015). Sendai Framework for Disaster Risk Reduction.
USAID (2016). Data Disaggregation by Geographic Location.
Vos, M., & Sullivan, H. T. (2014). Guest Editors’ Introduction: Community Resilience in Crises: Technology and Social Media Enablers. Human Technology, 10(2).
Wall, I., & Chery, Y. G. (2011). Ann Kite Yo Pale: Let Them Speak: Best Practice and Lessons Learned in Communication with Disaster Affected Communities: Haiti 2010. Last updated November.
Web Foundation (2015). Women’s Rights Online.
Xu, J., Nyerges, T. L., & Nie, G. (2014). Modeling and representation for earthquake emergency response knowledge: perspective for working with geo-ontology. International Journal of Geographical Information Science, 28(1), 185-205.
Yuan, W., Guan, D., Huh, E. N., & Lee, S. (2013). Harness human sensor networks for situational awareness in disaster reliefs: a survey. IETE Technical Review, 30(3), 240-247.