Bureau of Intelligence and Research

U.S. Department of State

Humanitarian Information Unit


Workshop on Complex Emergencies: Collecting Data, Managing Information, Seeking Knowledge


System Planning Corporation

USA Today Building

1000 Wilson Boulevard, 31st Floor

Rosslyn (Arlington), Virginia


Monday, September 8 – Tuesday, September 9, 2003



I.                    Summaries of Remarks made by Presenters and Panelists, September 8

II.                 Summary of Remarks of Luncheon Speaker, Ms. Julia Taft

III.               Notes from the three Breakout Groups

IV.              Presentations of Recommendations from the Breakout Groups, September 9

V.                 Key Recommendations and Action Items from the internal USG discussion

VI.              Workshop Presenters, Panelists and Participants List




I.  Summaries of Remarks made by Presenters and Panelists


September 8, 2003:  Opening Remarks


William B. Wood, Deputy Assistant Secretary, Bureau of Intelligence and Research, U.S. Department of State


There are many lessons that can be learned from complex emergencies, especially recent and current activities in Iraq, that relate to the need for better information management. We need to overcome the interagency turf battles, intransigence and the lack of flexibility and innovation that plague our bureaucracies.  We need to do a better job at making our data, information and knowledge more useful and accessible for policy and decision making.   


Despite earlier efforts, the formulation of an Iraq Reconstruction Information Strategy and the creation of a dedicated Information Management Unit is just now being designed and implemented.  This strategy and unit will be addressing the key challenges outlined for this workshop: 1) improving ways of collecting, assessing and analyzing data and information from the field; 2) improving coordination, collaboration and information sharing among various organizations; and 3) providing products and services that enhance knowledge, understanding, and decision making.


While I was in Baghdad, the terrorist bombing of the UN headquarters took place, killing Sergio Vieira de Mello, several staff members of the UN Humanitarian Information Center, and many other brave humanitarian workers.  We owe it to them to improve our information capabilities, which will make us more effective in helping those in need.     


Panel One: Views from inside the U.S. Government

William Garvelink, Deputy Assistant Administrator, Bureau for Democracy, Conflict and Humanitarian Assistance, U.S. Agency for International Development


Complex emergencies have gotten much more complex. Senior officials need different types of information than personnel in the field.  Information is needed so senior officials understand the crisis, but they also need information so that they can brief other decision makers, Congress and the media.  Sometimes this information is more general and anecdotal than that you would want to use in the field. 


Decision makers need a basic snapshot of information on the overall political and humanitarian situation.  Progress on humanitarian response requires mortality data, information on population movements around the country and outside, and a value judgment as to whether we are meeting the critical needs.  In complex emergencies, human rights information and security information is also needed.  They need to know what agencies are doing, how much they’re spending, and what and how much money other donors are spending on activities.  Decision makers need objective information on impending problems and issues so there are no surprises.


The HIU is a good organization to deal with this side of the equation, to provide a template of information and common situational awareness that all agencies in the USG can use.  Otherwise, without a common set of information, confusion arises over terminology and the situation.  We could make a stronger case for requirements from other donors.


Arthur “Gene” Dewey, Assistant Secretary of State, Bureau of Population, Refugees and Migration, U.S. Department of State


Managing complex emergencies requires successfully managing information. Two not-too-recent experiences are examples of success stories in information management: the Organization for Emergency Operations in Africa in 1984-85 and the UN Transition Group for Namibia.  Two examples of failures in information are the 1984 measles outbreak in refugee camps in eastern Sudan and the 1994 Rwanda genocide and refugee crisis in Zaire.  These two examples demonstrate that failures in managing information can result in vast numbers of unnecessary deaths. 


There are several lessons identified (but still not learned) with respect to information management.  The first is the need to discipline the system, not to be subjugated by it.  The second is the need for a better understanding of early warning indicators -- crisis managers don’t know what they mean.  There is no system in our government that forces policymakers to confront these indicators in a realistic way. A third lesson is the need for information for planning and pre-deployment training, which are critical for developing a coordinated response.  


Another lesson is the need to establish a common operating picture that reflects the activities of and can be used by all involved players. The need for a common operating picture occurs when the national inter-agency information is conflicted, which happens continually in Iraq.   Finally, information operations must be harnessed to drive realistic transformation over the civil-military force, not just the military.  That was done well in Afghanistan but was not repeated exactly for Iraq.  In Afghanistan, US CENTCOM Commander General Tommy Franks was able to refer humanitarian assistance to groups who he acknowledged could do the job much better and freed him to go about his military operations.


LCDR David Tarantino, Stability Operations, Office of the Secretary of Defense, U.S. Department of Defense(DOD)


The DOD has a role in humanitarian assistance and complex emergencies on a day- to -day basis.  Information is critical because it guides decision making.  The information needs are insatiable and DOD must distinguish between data and information.  This is doable but it is difficult – it requires a mindset that is different from dealing with emergencies in a traditional way.  Information needs to be thought of as a valuable commodity, it needs to be pre-positioned and packaged to be used effectively.


International organizations and NGOs, the eyes and ears on the ground, are critical to DOD.  However, it is a two way street – it is important for DOD to declassify as much as possible so that the information can be useful to those on the ground.  Information exchange depends on this flow of declassified data. 


We have to start small to build information protocols.  We need to improve the collection and assessment process and make this part of interagency exercises.  Central Command has learned a lot of lessons – the military and civilian sectors know each other now, and while they may disagree, at least they are in communication.  This civilian-military collaboration needs to be repeated at European Command and Pacific Command.  In Central Command, people move on – there is a two-year rotation.  As such, operations need to be institutionalized.  DOD wants to cooperate to improve coordination and collaboration at the field level and the Washington level. 


Panel 2:  Views from outside the U.S. Government


H. Roy Williams, President and CEO, Center for Humanitarian Cooperation


Unfortunately, in the realm of humanitarian emergency management, information management is usually an afterthought.  It is usually a matter of convenience of the day rather than a part of the system.  There are many reasons for this.  One major one is staff turnover.  People are not generally trained in information management.  Opportunities for training are very brief and much of it is centered on security.   Another reason is increasing compartmentalization. The humanitarian structure, which is becoming increasingly inclusive, does not have a process in which it talks to each other in terms of information.  Information dissemination begins in the field, but it is very informal.  The relationships among the organizations at the headquarters level have to be much firmer in terms of information management and exchange. 


NGOs don’t like to share information with the military, which is an issue.  Likewise, the military clearly has useful information but is usually not inclined to share.  This does not encourage open-mindedness and transparency among the other participants.  In Afghanistan, institutionalization of a system for sharing information in the field and at headquarters between the military and civilian organizations was occurring.  This happens very rarely.  The U.N and NGOs were invited and sent representatives to serve as information liaisons at US CENTCOM in Tampa.  This creates a process of organization and cooperation among all those involved.  It is not so much what you share but how the organizations see what you share.


This is a system problem.  Information needs to be viewed as a process, not simply a checklist of accomplished actions. The process in and of itself could change the way structures operate and the assumptions they make about their behavior.  The behavior can restrict the way a product is developed. The only way is to establish a standing system of communication and discussion about what we are all about.



Sue Lautze, Director, Livelihoods Initiative Center, Feinstein Famine Center, Tufts University


What is the problem we’re trying to solve?  What are the assumptions lying behind the three questions facing us today?  The point of departure is to ask, “Is there something critically wrong with information data systems?”  The systems are working very well for many actors.  It is a functional dysfunction.  A different point of departure is to ask “For whom are these systems not working?”  It could be said that the current information and data systems are failing the people suffering from complex emergencies far more than the institutions dealing with them.


Another assumption is that data, information and knowledge are interchangeable.  But the questions we are asking are related to data and information, not knowledge.  There is considerably less interest in knowledge, which requires deeper levels of engagement.  Information is about facts rather than processes of understanding.  We are awash in information.  Data and information must be analyzed to understand the symptoms of the problems.  This implies delving into the realm of knowledge.


We need to understand the ever changing nature of emergencies.  In complex emergencies, the most critical problem is translating data and information into knowledge and understanding the processes and vulnerabilities driving the symptoms of the problems. For example, the relationship between violence and vulnerability must be better understood.  The humanitarian community has yet to shed its anti-institutional bias and come to understand the linkages between global processes and local issues, which often generate the violence that characterizes complex emergencies. 


Jeffrey Henigson, Manager, Map Center, United Nations Office for the Coordination of Humanitarian Affairs (OCHA)


OCHA has tried to improve information services to the international humanitarian community through two projects: ReliefWeb and establishment of Humanitarian Information Centers (HICs).   ReliefWeb, established in 1996, is a global humanitarian information website, while there are currently seven HICs based in field locations providing operationally relevant information to humanitarian aid workers. 


The HIC provide many services to the humanitarian community in the field, including technical support, survey design, data archives and a help desk service.  The HICs also provide valuable products – GIS maps, data bases, meeting schedules, and other coordination tools. The HICs promote data standards that facilitate the collection, dissemination and analysis of information in the field. .


There are many challenges associated with both global and field humanitarian information management and exchange. One is ensuring quality, timeliness and consistency of coverage. Another is managing information overload.  For the HICs, there is a challenge of getting information professional staff deployed to the field, especially given the security concerns and the bureaucratic red tape..


II. Summary of Remarks by Ms. Julia Taft, Assistant Administrator and Director, United Nations Development Programme


Information is needed through all phases of an emergency situation – prior to the emergency, during the emergency and in the post-emergency phases of reconstruction and development.  In Afghanistan, data and information were being collected by UNDP and FAO on agriculture, climate, economic livelihoods, etc.  This data and information became the baseline foundation for the Afghanistan HIC.  During the height of the emergency, the priority shifted to tracking food deliveries, movement of vulnerable populations, location of humanitarian organizations, accounting of aid contributions, etc.  However, baseline data and indicators still need to be collected for the resumption of the development phase.


In Nepal, a UNDP GIS mapping project helped show the geographic, historical and economic causes of the Maoist insurgency. Information is critical for identifying vulnerabilities.  In southern Africa, there was lack of reliable data on HIV/AIDS and its impact of the society.  Numerous organizations were collecting data and conducting assessments, but there was little harmonization of this information.  


Information needs to benefit and be owned by the people of the country. The more information is seen as being for the benefit and service of the international community instead of the people, the less information can serve the real needs of the population and long-term development interests.  We can not go in and say we know what is best for the people and that we have all the information. We need to involve the grassroots, local governments and indigenous leaders in training and information systems.   


III. Notes from the three Breakout Groups


Breakout Group 1: How do we improve the collection, assessment and analysis of humanitarian-related data and information from the field?


Participants raised four main issues with respect to availability and usability of humanitarian-related information.  These included:


1) A lack of pre-crisis data (especially historical, time-series data) limits ability to assess current situation. 

– No context for assessment of current situation

– No baseline for trend analysis

– Limited institutional capacity (local) to provide access to existing data or for continued collection of time-series data



2) Lack of standardization of data specifications and collection methods limits usability:

– Collection/measurement process can bias results

•e.g., cultural/language differences; acceptance of technology

– May not be comparable across time and space

•e.g., the same question asked in two different countries can produce different results; how to compare infant mortality as reported by women of childbearing age?

– May not actually measure outcomes                        

•e.g., # of medical facilities vs. access to medical facilities



3) Information needs, and the ability to collect it, varies over time

– Phase of emergency/crisis … different information is required before/during/after complex emergency

– Security … very difficult to collect this information during hostilities


4) Limited incentives for sharing data

– Even may be disincentives for reporting positive results  … “Progress has been made; funding is shifted to more pressing problems.”


Next, attention was turned to “What is currently working?  Participants pointed out that:


1) Existing data standards / frameworks offer starting points for data

– SPHERE project specifies minimum standards for provision of disaster assistance, especially in nutrition/health care areas … make sure that data collection is linked to these standards, for example

– SMART advocates development of a standardized methodology, shared global database and training systems to collect mortality and nutrition measures

– SHARE advocates documentation of sourcing, dating, and geo-referencing of field-collected data (e.g., metadata)


2) Data collection templates (i.e., establishment of minimum content) do help with data consistency.  However the following caveats apply:

– They must be targeted (i.e., don’t turn it into a 10 page “quick” form) and flexible to incorporate free-text notes

– They only provide limited opportunity for local involvement in design/testing/refinement; hard to avoid cultural bias without such involvement.  Strongly suggest at least a review of the template by locals before wide-scale dissemination.

– They only provide limited ability to get to “knowledge” … they mostly collect “facts” … how many wells are functioning, how many calories are people consuming, etc.  Not “what is the impact of this information.”


3) Tools/technology not perceived to be limiting

– PDA; GPS; GIS; Satellite Communications  … when available, these work well to automate collection

Training on the use of these technologies must, however, be provided prior to going into the field. 


Finally, the group made several recommendations for improving the usability (especially by parties other than the collecting agency) of field-collected data:


1) Coordinate data collection plan between agencies (i.e., develop a common understanding of what is to be collected, with what methods and by whom)

– Coordinate content; terminology; methodology

– Establish data sharing agreements before data are collected if possible


2) Encourage donors to require data sharing and core data/metadata content


3) Actively develop and promote training

–Data templates/collection methodology

–Develop in-country data maintenance capacity

  4) Develop Humanitarian Assistance data model/standards

–Sector-based templates for core/minimum content (data and metadata)

–Local dialog to test/refine specific language and collection methods (be culturally sensitive)


Breakout Group 2: How can we improve inter-agency and international information coordination and better utilize collaborative tools?


Participants raised the following issues related to this topic:

What is working?

Humanitarian Information Centers are accomplishing sharing at the field level


How can the current situation be improved?

Making classified information available

            - -Pre crisis

                        -Understanding situation

                        -Existing capacity

                        -Early warning systems (commercially available)

                        -Simulation (modeling – Sim City)

                                    -good idea but need to be careful how the results are used

                                    -POST application at Cobra Gold

                                    -point of departure


                        -Internal methods that develop during crisis handling



                        -Handoff to development organizations



Breakout Group #3: What are the various information needs, and what are the products and services that should be provided during a complex emergency humanitarian response?


Breakout Group 3 approached answering this question by first stipulating that it would view the problem through the eyes of US government policy makers and program managers in Humanitarian Assistance arena.   It then proceeded to develop answers in three areas:  information needs, products and services needs, and enabling activities. 



A.  Information Needs




B.  Product & Services




C.  Enabling Activities




IV. Presentation of Recommendations from the Breakout Groups


1.  Conduct Data/Information Inventory and Gap Analysis

·        What agencies are involved/institutional capacity to analyze data

·        Capture expertise

·        Information needs assessment

·        What data/information/services/knowledge resources are available

·        Where they can get the data/information/sit reps etc

2.  Emphasize Cross-Institutional Coordination and Data/Information Sharing

·        USG-supported working groups/steering committees (GIST)

·        Need an NSPD? US govt. framework for data/information sharing

·        Study extant policies/laws facilitating/impeding data/information sharing

·        Recommend lead agency for information coordination

·        Encourage Donors to Require Data Sharing and Minimum Data Content

3.  Coordinate Data Collection and Sharing Strategies and Plans Between Agencies

·        Protocols, MOUs (good examples?) -- data/information sharing

·        Making De-Classified Data/Information Available

·        Agreement on Standards and Protocols (Data, Information)

4.  Develop Common Trend Analyses and Presentation Methods

·        Find out what is being done now (academia)

·        Develop trend analysis

·        Develop/apply decision making capacity (based on existing humanitarian standards)

·        Make Better Use of Advanced Analytical Tools

·        Improve knowledge visualization for decision makers

5.  Use Collaboration to Get Buy In – Working Groups

·        Customize the Collaboration Tools to the Area

·        Establish communities of interest

6.   Develop Multi-Sector Training Scenarios for Away Teams

·        Create training materials (tutorials, primers, etc)

·        Identify short term training/long term education

·        Consider accreditation based on training

·        Implement Best Practices Reach-Back Capability

7.  Develop Integrated Situation Awareness

·        Conduct information inventory  -- need that is being served

·        Evaluate ways to consolidate/integrate/hyperlink summaries and target analysis

8.   Better Continuity of Transition – Transitional Processes

·        Identify/Develop In-region information capacity building

·        Pre-identify existing resources/capacity (in-region, US, NGOs, academic, commercial)

·        Link to information transition strategy


V. Key Recommendations and Action Items from USG Discussions

1)      Conduct Data/Information Inventory and Gap Analysis

a)      What agencies are involved/institutional capacities to analyze data?

­       Who is doing a data/information inventory? (SDR – Larry R.)

­       SDR (Subcommittee on Disaster Reduction)

­       Parallel activities – S&T agreements addressing disasters from SDR – precedence?

­       HIU – could maintain a database on data/information/report resources

b)      Capture expertise

c)      Information needs assessment

d)      What data/information/services/knowledge resources are available?

e)      Where they can get the data/information/sit reps etc


·               HIU identify top 20 countries (Leslie C./HIU)

·               Based on top 20 countries of interest, get inputs from each USG agency on types of data/information/reports (Larry R./SDR; Leslie C./HIU)

·               HIU to set up a design concept for an accessible database (Kathleen M/HIU)


2)      Promote Inter Agency Coordination and Data/Information Sharing

a)      USG-supported working groups/steering committees (GIST)

b)      Study extant policies/laws facilitating/impeding data/information sharing

c)      Support ongoing efforts to improve/encourage Data Sharing and Minimum (core) Data Content

­       Inter Agency Standing Committee (UN)

d)      Protocols, MOUs (good examples?) -- data/information sharing

e)      Agreement on Standards and Protocols (Data, Information)


·               Set up a inter agency working group on information sharing among USG agencies (focus on preparedness, contingency, selected priority countries, information sharing) (Dennis K./HIU)

·               Working Group outline an action plan based on a review of recommendations (covering #2) Review recommendation, identify and prioritize actions and assign leads. (Dennis K./HIU) 


3)      Review releasability of Data/Information (limited distribution, proprietary, derived products)


·               Working group to work on releasability/collection recommendations for sharing data (Alan Huguley/NIMA)


4)      Develop Common Trend Analyses and Presentation Methods

a)      Find out what is being done now (academia)

b)      Develop trend analysis

c)      Develop/apply decision-making capacity (based on existing humanitarian standards)

d)      Make Better Use of Advanced Analytical Tools

e)      Improve knowledge visualization for decision makers


·               Layout an approach/provide recommendations (Craig C./PDC; Dave H./COE)


5)      Use Collaboration to Get Buy In – Working Groups

a)      Customize the Collaboration Tools to the Area

b)      Establish communities of interest

c)      Practice Data/Product “Diplomacy “ and Outreach


·               Evaluation of collaborative tools for Iraq (Leslie C./HIU)