GUIDANCE NOTE

Household and
Building Damage
Assessment

December 2019
Table of Contents
_Toc28010279
Table of Contents ................................................................................................................................................. 1
1. HBDA Key Actions and Relevant Sections to Read .................................................................................... 3
2. What is the Household and Building Damage Assessment (HBDA)? ........................................................ 5
A. What is the HBDA? ................................................................................................................................... 5
B. The HBDA Toolkit ..................................................................................................................................... 5
C. The HBDA in the Assessment Cycle & Decision-Making............................................................................. 6
D. Limitations: Potential Issues to Bear in Mind .............................................................................................. 7
3. Preparation for the HBDA .......................................................................................................................... 10
A. Rapid Context Evaluation ........................................................................................................................ 10
B. Assessment Working Group, Core Team, & Coordination......................................................................... 10
C. HBDA Core Team & Resources............................................................................................................... 11
D. Model Work Plan..................................................................................................................................... 13
4. Secondary Data Review............................................................................................................................. 15
A. Objectives ............................................................................................................................................... 15
B. Conducting the SDR ............................................................................................................................... 15
C. Type of Information and Sources for the HBDA ........................................................................................ 16
5. Methodology & Assessment Design ......................................................................................................... 19
A.Methodology and Data Analysis Plan ....................................................................................................... 19
B.Sampling Strategy ................................................................................................................................... 21
C.Adaptation of the Data Collection Tool ..................................................................................................... 24
6. Data Collection .......................................................................................................................................... 26
A. Preparation for Data Collection & Enumerators’ Training .......................................................................... 26
B. Day-to-Day Management ........................................................................................................................ 27
C. Mobile Data Collection and Kobo Toolbox ................................................................................................ 28
7. Data Analysis ............................................................................................................................................. 29
A. Data Cleaning and Processing ................................................................................................................ 29
B. Data Analysis ......................................................................................................................................... 30
C. Analysis Tools for Visualization................................................................................................................ 32
D. Information Products, Target Audience, & Dissemination ......................................................................... 33
Annex 1: Definitions ........................................................................................................................................... 35
Annex 2: Reference Documents ......................................................................................................................... 37
Annex 3: Information Management, Visualization, Mapping and Navigation Tools .......................................... 38
Annex 4: Terms of References (ToR) of the HBDA Team .................................................................................. 39
ToR, Program Lead/ Assessment Coordinator ................................................................................................... 39
ToR, Assessment Officer .................................................................................................................................. 39
ToR, Information Management/ GIS Officer/ Remote Sensing Analyst ................................................................ 39
Annex 5: Templates ............................................................................................................................................ 40
Data Cleaning Logbook .................................................................................................................................... 40
Data Management Plan .................................................................................................................................... 40

1 1
 Methodology Note ............................................................................................................................................ 42
Annex 6: HBDA Hardware and Software Requirements .................................................................................... 45
Annex 7: UNDP-UNOSAT Standard Operating Procedures ............................................................................... 48

2 2
1. HBDA Key Actions and Relevant Sections to Read

This summary outlines key actions to be undertaken in the Household Building Damage Assessment (HBDA). It can
serve as a checklist for steps and considerations for conducting a HBDA, as well as direct readers to the appropriate
sections of the handbook for each action. The full handbook provides theoretical foundation as well as practical
guidance, tools and resources for practitioners.

Action Handbook Section

1. Decide whether a HBDA may be appropriate
Understand what the HBDA can do and what is available in the HBDA 1A. What is the HBDA?
toolkit. 1B. The HBDA Toolkit
1D. Limitations: Things to Bear in
Mind
Decide whether it is appropriate to launch a HBDA, based on information 1C. The HBDA in the Assessment
needs, resources, and how it fits within the humanitarian response. Cycle and Decision-Making
1D. Limitations: Things to Bear in
Mind
Conduct an initial review of basic information about the crisis, to help 2A. Rapid Context Evaluation
determine scope and feasibility of the HBDA.

Advocate with the government and international actors for a HBDA, which 2A. Rapid Context Evaluation
should be conducted upon the request of the government.

2. Prepare to conduct a HBDA

Establish the HBDA Core Team (with full involvement of relevant UNDP 2C. HBDA Core Team & Resources
country office colleagues), which will run the HBDA on a day-to-day basis. 2B. Assessment Working Group,
Core Team & Coordination
Annex 4: Terms of References for
the HBDA Team
Work with relevant stakeholders through an Assessment Working Group 2B. Assessment Working Group,
(AWG). Ensure all AWG members are trained on the HBDA toolkit. Core Team & Coordination

Organize the stakeholders to contribute to the HBDA with human, data, 2B. Assessment Working Group,
and financial resources. Core Team & Coordination

Create a workplan to estimate timeframe of the HBDA from inception to 2D. Model Work Plan
dissemination of findings, as well as assign tasks.

Conduct an in-depth Secondary Data Review (SDR), including collating 3A. SDR Objectives
and analyzing information on pre-existing damage as well as already 3B. Conducting the SDR
collected in-crisis data. This should clearly identify information gaps and 3C. Type of Information and
provide a baseline of information for comparison with HBDA data. Sources for the HBDA
Annex 7: UNDP-UNOSAT SOPs
Clarify and agree the final HBDA objectives and use of its data for 3A. SDR Objectives
humanitarian and recovery, including the relationship with other 1C. The HBDA in the Assessment
assessment processes (e.g. MIRA, PDNA etc.), and how it contributes to Cycle and Decision-Making
recovery frameworks, donor requirements, strategic planning.

3 3
 3. Design the HBDA
Decide on the HBDA coverage and methodology (including sampling 4A. Methodology and Data Analysis
strategy and research questions). The scope and sampling strategy can be Plan
selected based on the four scenarios developed for the HBDA which 4B. Sampling strategy
depend on available resources and objective of the assessment. Annex 5: Templates

Select appropriate indicators and questions from the HBDA library of C. Adaptation of the Data Collection
questions. Adapt any context-specific indicators. Tool
HBDA library of questions
Adapt and test the questionnaire using Kobo Toolbox and the HBDA 5C. Mobile Data Collection and
templates, including selecting a server and setting up a dedicated account. Kobo Toolbox
Ensure the questionnaire if reviewed by stakeholders.

Make sure the above are outlined in the following documents: (1) 4A. Methodology and Data Analysis
Methodology, (2) Data Analysis Plan, and (3) Data Management Plan. Plan

4. Collect HBDA data

Complete any HR, administrative, logistical and related tasks for data 5A. Preparation for Data Collection
collection (e.g. hiring enumerators, preparing equipment and training & Enumerators’ Training
venues).

Ensure that Assessment Officers, GIS Officers are proficient in the use of 5C. Mobile Data Collection and
Kobo Toolbox and are able to train enumerators. Kobo Toolbox

Train enumerators on mobile data collection, including general principles, 5A. Preparation for Data Collection
practical aspects of the HBDA and the data collection questionnaire. & Enumerators’ Training

Launch data collection and manage this on a day-to-day basis, keeping 5B. Day-to-Day Management
track of progress on an ongoing basis.

5. Analyze HBDA data and share findings

Clean data on a daily basis according to defined cleaning protocols. At the 6A. Data Cleaning and Processing
end of data collection, conduct a comprehensive final review of the dataset.

Conduct the four phases of data analysis: description, explanation, 6B. Data Analysis
interpretation / operationalization, and early warning & preparedness.

Visualize findings using Microsoft Power BI, and adapt existing dashboard 6C. Analysis Tools for Visualization
templates. Consider other conventional GIS platforms to further analysis.

Compile findings into other information products as requested / relevant. 6B. Data Analysis
This may include a final report, datasets, factsheets and profiles, 6D. Information Products, Target
dashboards, infographics and maps. Audience & Dissemination
Annex 3: IM, Visualization, Mapping
and Navigation Tools
Share and disseminate findings and information products in collaboration 6D. Information Products, Target
with the government. Audience & Dissemination

4 4
2. What is the Household and Building Damage Assessment (HBDA)?

A. What is the HBDA?

Natural disasters and conflicts affect buildings and infrastructures, e.g. residential houses, schools, hospitals, police
stations, and businesses, which in turn has an impact on local communities. In this context, one of the most challenging
and crucial components of the recovery process is the rehabilitation of these structures based on reliable data on the
level of damage. In addition, humanitarian crises result in significant needs among affected populations in others
humanitarian sectors, including Health, Food Security, Education, and Water, Sanitation, and Hygiene (WASH).
The United Nations Development Programme (UNDP) with the support of IMPACT Initiatives has developed
this Household and Building Damage Assessment (HBDA) toolkit to facilitate the assessment of buildings and
infrastructure affected by crises (http://www.undp.org/content/buildingdamageassessment/en/home.html), as well
as assess a series of basic socio-economic indicators in Health, Food Security, Education, and WASH at the
household level. The HBDA allows prompt collection of digital and georeferenced data on damage to infrastructures
and residential and commercial buildings, household vulnerabilities, utility services, and needed materials for repairs.
The toolkit is designed to be flexible and if started early, can produce valuable information for the Post Disaster Needs
Assessment (PDNA)12, especially for the housing, community infrastructure and social sectors.
The HBDA is a comprehensive and multiphase process that involves mobile data collection through enumerators,
followed by data analysis, visualization and dissemination for use by national and local authorities, humanitarian and
development partners. The library of questions within the HBDA is designed to be adaptable to different contexts. It
includes a priority rating per indicator that enables flexible management of the length of the data collection tool,
depending on available resources and government priorities. The analysis produced by the HBDA aims to inform
government prioritization of relief and recovery resources, serve for strategic and operational humanitarian and
recovery planning, as well as produce evidence for raising awareness and advocacy (see Part 6).
This Guidance Note presents the tools necessary for the implementation of the HBDA, including the HBDA
toolkit, the Kobo Toolbox (https://www.kobotoolbox.org/), and the analysis and visualization application
Microsoft Power BI (https://powerbi.microsoft.com/fr-fr/). Additional applications to potentially enhance HBDA
capabilities and/or allow greater flexibility are listed in Annex 3.

B. The HBDA Toolkit

The HBDA relies on mobile data collection, using the Kobo Toolbox – an open-source platform that allows rapid
information gathering using smartphones or tablets and facilitates the post-collection data analysis. Once data
collection is complete, the data is cleaned, analyzed, and deployed to Microsoft Power BI, an intelligence platform for
data analysis and visualization. Data and associated analysis from the HBDA are also exportable to other platforms.
The HBDA data collection tool relies on a library of questions with seven modular sections: A – Building Location
(15 questions), B – Building Use and Household Information (138), C – Building Size (7), D – Building Materials (20),
E – Building Damage and Repairs (24), F – Building Services and Debris (30), and G – Building Preparedness (9). The
full library of questions has over 240 questions.

1
2
For more information on the PDNA, see UNDP, 2014

5 5
It should be noted that not all questions will be asked in each HBDA. The length of the data collection tool will
vary depending on available resources within the assessment team, the information needs, and the access situation
in the particular context. The core of the HBDA includes sections A, C, D, E, and F, which will likely be part of all
HBDAs.
Section B is a standalone household and infrastructure assessment that allows collection of granular indicators at the
household and individual infrastructure (e.g. hospital, school) level. It is therefore by far the largest section. Section G
is the preparedness section, which may not be relevant in every context.
To simplify indicator selection, the library of questions includes a priority column that ranks each indicator on
a 1 to 3 scale. The indicators ranked 1 are considered the most vital while the indicators ranked 3 may be omitted if a
lighter assessment is conducted. It should be noted that this ranking is indicative and may change depending on
context. Specific ways in which the data collection tool may be adapted are explained in Part 4.C but may for instance
include expanding the data collection tool with more sectoral questions.

C. The HBDA in the Assessment Cycle & Decision-Making

The HBDA should fit within the larger framework of the humanitarian response, deployed when:
• a new crisis emerges and results in building damage, or
• there is a sudden or substantial change in an existing crisis which leads to building damage.
Conducting the HBDA will help inform the response to a crisis if:
• additional information on damage to buildings is required for operational or strategic decision-making, or
• contingency planning is undertaken
The HBDA may vary in depth, research method, time frame, and cost. It is thus a flexible and adaptable assessment
that can be deployed at different moments in a crisis (see Figure 1).
Figure 1: HBDA Timing

6 6
The assessment team should keep this distinction in mind when launching the HBDA, as the phase of deployment of
the HBDA will impact on its objectives, length, and the indicators selected from the library of questions for the data
collection tool (see Part 4).
Table 1: Rapid and In-Depth HBDAs

Rapid HBDA In-Depth HBDA

Timing of • Rapid assessment phase (1-3 weeks) of a • In-depth assessment phase (>one month) of a
launch sudden onset disaster. sudden onset disaster.
• Ideally by the end of the search and rescue • Depending on context as necessary in conflict
phase. and complex crises, for example after a flare-up
in conflict intensity leading to significant damage.

Aims • Immediately build a robust evidence base to • In conflict and complex crises, collect data
inform decision making. geared towards recovery and durable solutions to
• Collect data on the most affected populations the crisis, developing a joint understanding with
and areas to inform the immediate response affected populations on long-term recovery
in shelter, livelihoods, access to basic needs.
services, rehabilitation of community • In later stages of a sudden onset disaster, collect
infrastructure, etc. granular data on affected populations and areas
• Inform the housing and infrastructure to inform response strategy and coordination.
sections of Post-Disaster Needs • If conducted at regular intervals, the HBDA may
Assessments that would normally start 2 become a monitoring tool for reconstruction,
weeks after a disaster (see for on the recovery and even preparedness.
assessment cycle, UNHCR, 2017).

Tools • Exclusive building damage assessment, • Comprehensive set of indicators from the library
omitting Section B from the library of of questions, likely to result in a longer
questions. It may only focus on identifying questionnaire, data collection period and
shelter needs soon after a disaster. resources required.

1Finally, the HBDA will always be deployed by UNDP in coordination with the national and local governments
as well as international and local humanitarian partners. In various cases, the HBDA may rely on resources
provided by partners, including enumerators and support on analysis to make it a joint needs assessment. In this
regard, the HBDA’s library of questions and methodology use agreed standards, including geographic tagging and
harmonized sectoral indicators, to allow for cross-analysis with other data collection exercises (see Annex 1 for
definitions and see for reference on types of needs assessments, UNHCR, 2017; and on coordinated and joint
assessment, Inter-Agency Standing Committee, 2012).

D. Limitations: Potential Issues to Bear in Mind

• Enumerators/ Data Collection Tool: A typical challenge in damage assessments is the need to ensure
appropriate, comprehensive, and standardized training for enumerators, in order to mitigate biases related to the
subjective perception of damage.
o It is important to make all possible efforts to have structural engineers join as enumerators, otherwise the
findings of the HBDA cannot be taken as a technical evaluation of the buildings’ structural integrity.
o Though it integrates some key sectoral indicators, the HBDA is not a true multi-sectoral assessment and
can only provide an overview of needs at the household level. It is recommended that people with

7 7
 expertise on the sectors covered by the HBDA are enrolled as enumerators and participate in designing
the methodology.
o It is important to enroll the appropriate number of enumerators. If there is a lack of enumerators, the
HBDA will take a longer time to complete.
• GPS Accuracy: The highest level of accuracy using mobile phones is at best 5m. However, this might not be
possible to obtain if the GPS signal is weak or interfered with (for example, if signal is obstructed by buildings).
This makes it nearly impossible to assign the collected GPS points for structures to their precise location on the
map in GIS, either using satellite imagery or Open Street Map. A unique ID is hence required.
o The issue of GPS readings registering with a delay and leading to incorrect detections may be mitigated
by setting a required GPS accuracy threshold or adding the obligation for the enumerator to register a
second GPS reading at the end of the assessment. If the enumerator’s phone is online, the accuracy of
the GPS reading improves.
o Note: It is possible to include a metadata field for the accuracy of each recorded GPS point.
• Mobile Data Collection: Kobo Toolbox allows for online and offline data collection on both the smartphones’
browser and the Kobo Collect Android application. Despite this, the HBDA suffers from the traditional limitations
of mobile data collection, including difficulty to manage qualitative data, possible stress for respondents in
answering through a mobile device, security risk for enumerator and respondent due to being seen with a mobile
device in some contexts, risk of the device being rendered inoperable (lost, stolen, out of battery), cost of procuring
and deploying the devices, and need for procurement of power banks.
o Ideally, the questionnaire should be fully finalized and approved by all stakeholders before starting data
collection with the Kobo platform. If changes to wording or options are made once data collection has
started, this can cause difficulties with the live dashboard feature within the Kobo platform.
• Connectivity: In some contexts, there may be extremely low bandwidth, or even worse, none. In such cases, it
may be worth considering using the raw suite of Open Data Kit (ODK) applications which Kobo it built on. ODK
Collect, an alternative Android data collection application, and ODK Briefcase, an offline data aggregator, can be
used in tandem in such cases with minimal changes to the HBDA workflow. These allow upload and download of
survey forms to mobile phones or tablets without requiring internet connection.
• Mapping: By default, Kobo Toolbox uses Open Street Map as an embedded base map. However, Open Street
Map is not up to date in every context.
o In order to capture every building and facilitate the work of enumerators, it may be possible to have Open
Street Map updated based on recent satellite or drone imagery. It is also possible to customize map data,
which can also work offline, depending on project needs.
o Further, there are additional ODK-based data collection applications geared specifically toward spatial
data collection, for instance GeoODK and Open Map Kit, which may present further opportunities for the
HBDA.
• Navigation: Navigation in some contexts can be difficult. There is no native navigation component to the Kobo
Toolbox. Therefore, it is often useful to use a separate application to navigate the areas to be assessed, as well
as organize enumeration teams and distribute work assignments.
o There are multiple applications that can support this, including Maps.me, SW Maps, MAPinr, which should
be evaluated further for their suitability for the HBDA.

8 8
• Unique Identification Numbers: Although during the pilot exercise Kobo Toolbox generated unique identification
(UID) numbers for buildings and households, further research is needed to develop the logic for HBDA UIDs.
These UIDs should allow for identification of different members of the same household living in different buildings,
as well as capturing information about multiple-household buildings.
• Data Rich vs. Data Poor Environments: The core HBDA methodology is adapted more toward data poor
environments, in which relatively little spatial data related to infrastructure exist, and GPS data must be collected.
However, in data rich environments, it may be possible to obtain detailed building information, including digitized
footprints and address registries. In such cases, attempts should be made to collect building damage data against
these datasets or databases using either existing unique building identifiers or new unique IDs that can allow the
datasets to be joined together reliably.
• Needs Assessments: Challenges in needs assessments include the need to ensure adequate sampling
strategies, train enumerators on avoiding leading questions, the risk of respondents consciously manipulating the
answers to orient the type and scale of aid, and the need to recruit specialized enumerators and have a gender
balance among them (see for reference on challenges in needs assessment, UNHCR, 2017).

Key Takeaways

• HBDA’s seven sections: A – Building Location (15 questions), B – Building Use and Household Information
(138), C – Building Size (7), D – Building Materials (20), E – Building Damage and Repairs (24), F – Building
Services and Debris (30 questions), and G – Building Preparedness (9).
• The Library of questions includes a priority column that ranks each indicator on a 1 to 3 scale, giving
an indication of its importance.
• HBDA’s data collection normally starts 1-4 weeks after a sudden-onset disaster during the rapid
assessment phase but can also be deployed later as an in-depth assessment and at regular intervals for
monitoring,
• Two major limitations to the HBDA to bear in mind:
o GPS accuracy using Kobo Toolbox is a problem
o Need for specialized enumerators e.g. structural engineers in assessment teams, to ensure that
findings can be used as technical evaluations of the buildings’ structural integrity

9 9
3. Preparation for the HBDA

A. Rapid Context Evaluation

In the HBDA, as in any assessment, there is a two-step secondary data review (SDR) – all the data that is
exterior to the HBDA – that needs to be conducted. The first step should be done quickly (1-2 days) at the
onset of the assessment process to determine the feasibility of the HBDA. Conversely, the second part of the
SDR is done during the data collection phase (see Part 3 and see for reference on context overview, UNHCR, 2017).
The first step of the SDR comes in parallel to reaching out to the government and other stakeholders, which is crucial
for obtaining data and information related to the context as well as ensuring buy-in of the process. This also forms the
first step of organizing stakeholders to contribute to the HBDA with human and financial resources. Advocacy with the
government and international actors can be facilitated through using the following tools from the HBDA toolkit: two-
pager on the HBDA, HBDA website, examples from previous assessments to demonstrate the added value provided
(including screenshots of Power BI dashboards, powerpoint presentations, factsheets etc).
The first part of the SDR is a review of basic information about the crisis. It involves quickly analyzing the existing
pre-crisis and in-crisis data collected by other actors, including humanitarian NGO and UN agencies, national or local
governments, civil society, faith-based organizations, satellite imagery (e.g. UNOSAT, Copernicus), crowdsourcing,
social media (e.g. QCRI social media analysis). The aim is to build an understanding of the context, of the crisis drivers,
and of the main information gaps. It is hence possible to determine whether the HBDA is needed given information
already available, as well as whether it is feasible given access constraints. It also helps identify which methodology is
possible, which HBDA components will be utilized to best address the information needs with the available time and
resources. It is important to conduct this process before launching the HBDA as it may well be that there is
either no need for a damage assessment, or that the situation is such that the HBDA is not possible due to
access constraints.

Checklist for the HBDA’s Feasibility in the Rapid Context Evaluation

For the HBDA to be relevant, timely, and needed after/ in a crisis, the following conditions need to be met:
✓ Access is possible to affected areas with regard to security and local authorities’ support for the HBDA
✓ The crisis has resulted in damage to buildings i.e. collecting information on shelter is a priority
✓ No other damage assessment is ongoing or has been completed in the last month/ since the outbreak of
the crisis. If so, evaluate the extent to which the HBDA would be able to collect relevant additional
information.
✓ Enough enumerators and resources (expert staff, funding, equipment for mobile data collection, etc.) is
available in UNDP to conduct the HBDA. If this is not the case, consider pooling resources with partners.

B. Assessment Working Group, Core Team, & Coordination

If the HBDA is both relevant and feasible, the next step is to
1. Establish the HBDA Core Team with full involvement of relevant UNDP country office colleagues (see
section C for details);

10 10
 2. Work with relevant stakeholders through an existing Assessment Working Group (AWG) or by
establishing one.
The AWG usually includes technical experts seconded from partner humanitarian organizations (UN agencies, ECHO,
INGOs, LNGOs, etc.) and local authorities (ministries, disaster management agencies, institute of statistics, structural
engineers, etc.). UNDP may co-lead the AWG with the relevant government counterpart.
Ideally, all AWG members should receive a training on the HBDA and familiarize themselves with the HBDA toolkit.
They should all be linked up through a WhatsApp group and other potential coordination channels (mailing list, Skype
group, Microsoft Teams, etc.).
The AWG should not duplicate existing assessment working groups and, to the extent possible, integrate into them in
order to pool resources for the HBDA and build synergies with other sectoral and multi-sectoral assessments, e.g.
MIRA and PDNA.
The AWG should prompt discussions with partners on methodologies and data collection tools and the development
of a joint analysis framework for the interpretation of findings.
Finally, the AWG should ensure that the HBDA Core Team, which runs the HBDA on a day-to-day basis, is regularly
liaising with relevant counterparts from other agencies, humanitarian clusters or sectors, and the local authorities.

The HBDA Core Team should engage with the AWG at all key stages of the HBDA process. At minimum, the
following points should be discussed with the AWG:
✓ Clear objectives on information needs, keeping in mind the purpose of the needs assessment
✓ Coverage (populations, sectors, and geographic areas)
✓ Type of decision-making that needs to be informed
✓ Timeframe for the start and end of data collection
✓ Methodology, including for data collection and analysis, data collection tools, analysis plan
✓ Coordination and division of responsibilities shown in an organigram
✓ List and mapping to engage stakeholders among national and local authorities
✓ Communication to all stakeholders, particularly the authorities, about the objectives and limitations of the
HBDA
✓ Ensure the link with PDNA and other relevant needs assessments with appropriate data, indicators and
timeframe

C. HBDA Core Team & Resources

The HBDA Core Team is part of the AWG and is in charge of carrying out the HBDA. Tasks include to set
objectives, identify necessary technical expertise, design the questionnaire, the data collection and visualization
platform, coordinate data collection and analysis activities, share findings to inform response planning and
programming, and report back to persons of concern.

11 11
The Core Team is a gender-balanced multifunctional team with sector experts and relevant technical specialists who
understand the local context. The HBDA Core Team Leader may also co-lead the AWG. The composition of the HBDA
Core Team may also change from one context to another depending on specific needs to be assessed. Team roles
can be filled by either national or international staff, with representation of national experts strongly encouraged to
ensure ownership and context sensitivity, among others. Aspects to be considered for team composition include
diversity (ethnical, gender) understanding of local political economy and stakeholders, local language and technical
expertise (e.g. fluency in the languages of the assessed areas, data literacy, understanding of technical indicators).
Table 2 below provides a list of the main roles in the HBDA Core Team. Team members may also fill more than one
role. Annex 4 provides the Terms of References (ToR).
Table 2: Description of Roles in the HBDA Core Team

o Lead and coordinate assessment teams

Team Lead/ o Liaise with external actors/ stakeholders to manage expectations around the HBDA
Assessment o Build consensus on the goals, information gaps, methodology and data analysis plan
Coordinator (AC) o Plan/ mobilize financial, logistics and human resources
o See Team Leader TOR in the annexes for more details
o Coordinate steps in the HBDA (methodology, SDR, data collection/ cleaning, analysis)
o Conduct access and context analysis, and integrate SDR information into the HBDA
Assessment Officer(s)
o Develop the data collection tools/ briefing and de-briefing enumerators
(AO)
o Coordinate the data collection process, including progress tracking and troubleshooting
o See Assessment Officer TOR in the annexes for more details
o Perform GIS/ Satellite imagery analysis
Information
o Support design of data collection tools, especially for mobile data collection, sampling
Management (IMO)/
frameworks, data storage systems, and analytical products
GIS Officer(s) (GISO)/
o Support analysis/ generating data visualization products, including dashboards, infographics,
Remote Sensing
and maps, especially using Power BI
Analyst (GISO)
o See IM / GIS Officer TOR in the annexes for more details
o Provide sector expertise for the indicators and analytical framework
Cluster Technical
o Conduct sectoral needs analysis and contribute to multi-sectoral analysis
Expert(s)/ Advisor(s)
o Check the collected data on damage by structural engineers and shelter experts
(CA)
o Brief/ de-brief with enumerators on technical indicators, including damages coding
In close collaboration with the AOs and UNDP CO management and operations manager
Admin (A)/ Finance
o Set in place and coordinate the assessment logistics
(F)/ Logistics (L) Staff
o Establish required human resources, ICT and other elements to carry out the HBDA
Liaison(s) (national o Conduct translations when needed for the data collection tools
staff) and Team o Support on context and access analysis, using local sources, social media, etc.
Leader(s) for o Coordinate the work of enumerators in the field
Enumerators’/ o Facilitate contacts with local and national authorities
Enumerators (national o Collect data in the field
staff) (E) o Support on context and access analysis

Table 3 below now provides a repartition of the main responsibilities in the HBDA Core Team.
Table 3: Responsibility Assignment Matrix of the HBDA Core Team for Key Tasks in the HBDA (A –
Accountable; R – Responsible; C – Consulted; I – Informed)

IMO/
AC AO(s) CA A/L/A ETL E
GSO

12 12
 Context Evaluation A/R C C C I C I
AWG/ Stakeholders A/R C C C C I I
SDR C A/R C R I C I
Methodology A R C R I I I
Enumerators'
Training C A/R C R C I I
Data Collection A R C C C I I
Data Cleaning C A/R C R I I I
Data Analysis C A/R C R I I I
Information Products A R C R I I I
Dissemination A/R R C R I I I

Deciding to conduct the HBDA has budgetary implications. Any HBDA budget should include an estimate of
supplies, personnel, training, transportation, communications, and security requirements. The cost of the HBDA should
be proportionate to the scale and scope of the crisis and expected benefits for the population of concern. If the HBDA
involves resources provided by partners, decisions have to be taken on how to pool and utilize them. Due consideration
should be given to the use of existing resources within communities of concern and ensuring whenever possible that
persons of concern are engaged in planning and implementing assessments.
Since the HBDA involves mobile data collection, attention should be given to the equipment for enumerators, including
in providing smartphones/ tablets, GPS devices, power banks and identifying t-shirts, vests, and caps. In damage
assessments, enumerators may need to be provided with measuring tapes/ meters.
Enumerators may be employees from UNDP, UN, national and international organizations and government. These
institutions can also recruit enumerators to carry out the assessment and mobilize volunteers (e.g. university students,
private citizens, etc.).

D. Model Work Plan

Table 4 provides an estimated number of days for the steps of the HBDA from inception to dissemination of findings.
Some of these steps are likely to be conducted in parallel by different members of the HBDA Core Team. Overall, the
HBDA should take between 3 and 6 weeks depending on its scale and methodology.
o Context evaluation: 1-2 days
o AWG Set-up and Stakeholder Engagement: 2-4 days
o SDR: 3 days and then monitoring until then end of the HBDA
o Methodology Development: 2 days
o Enumerators training: 1-3 days
o Data Collection: 1-2 weeks
o Data Cleaning: 1-2 days at the start of data collection, ongoing throughout the data collection process, and 2
days at the end
o Data Analysis: 4 days-1 week
o Report, Maps, and Other Information Products Production: 4 days-1 week
o Findings Dissemination: 1-3 days after end of data analysis but is likely to start earlier

13 13
Table 4: Indicative Work Plan for the HBDA (dark grey: minimum time; light gray: maximum time)

5th -
HBDA Weeks 1st 2nd 3rd 4th
6th
4-
(days) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3
7
Context
Evaluation
AWG Set Up
SDR Ongoing
Methodology
Training
Data Collection
Data Cleaning
Data Analysis
Information
Products
Dissemination

Key Takeaways
• Conduct a rapid context evaluation to assess the feasibility of the HBDA
• Establish the Assessment Working Group (AWG) that develops the plan for the HBDA, liaises with
stakeholders, and coordinates the assessment team
• The assessment team includes a Program/ Assessment Lead; Assessment Officer(s); Cluster Technical
Expert(s); Information Management/GIS Officer(s); Admin/Logistics/Finance Staff; Team Leaders for
Enumerators; Enumerators
• The completion of HBDA should take between 3 and 6 weeks

14 14
4. Secondary Data Review

A. Objectives
The secondary data review (SDR) is a rigorous process of collating existing data and information on the
affected population and areas of focus of the HBDA synthesizing this information, and conducting analysis
(see Tables 4 & 5). In line with the HBDA design and methodology, the SDR establishes what is known and unknown
about the situation and its impact. It is then possible to identify information gaps by comparing the existing consolidated
data to the information needs that were agreed in the data analysis plan (see Part 4.A). For more information on the
SDR process see DEEP Secondary Data Review Basics.
For damage assessments in natural disasters, the SDR includes the aim to collect information on the
magnitude of the phenomenon, including wind speed, path, and level of precipitation for cyclones, epicenter location
and depth, aftershocks, and potential for tsunami, landslides, for earthquakes, speed of the lava flow and level of
airborne ash in volcanic eruptions, and other information in other natural disasters. In armed conflicts, the SDR aims
to gather information on the intensity of the fighting, especially of indirect and aerial fire.
For the HBDA, it is particularly important to have in-crisis information available through satellite imagery. Thus,
specific techniques may be used to assess damages in hard-to-reach areas or immediately after the crisis before the
HBDA is deployed, including remote sensing for in-depth damage assessment and determination of other visual
characteristics of geographical areas of interests. Remote sensing requires appropriate satellite imagery and/ or aerial
photographs, combined with expertise in imagery analysis and knowledge of the ground situation (see for reference
on remote sensing and satellite imagery best practices, UN-Spider Portal, 2019; CIEDRS, 2003; Copernicus, 2019).
UNDP relies on support from the UNOSAT for immediate crisis response (see Annex 7).
The SDR is therefore a key component of the HBDA. It ensures that all available secondary data, including on pre-
existing damages to buildings and already collected in-crisis data, is used before investments are made to collect
primary data. It moreover contextualizes the HBDA primary data by providing a baseline.
An SDR should be updated throughout the humanitarian crisis as more information about the crisis becomes available,
strengthening situation awareness among humanitarian actors. It is recommended that the AWG maintain an updated
secondary data review throughout the HBDA and a registry of completed and ongoing data collection exercises
(sometimes called a survey of surveys (SOS)) (see for reference on assessment registry, ACAPS, 2011).

B. Conducting the SDR

To undertake an SDR in a coordinated manner, a best practice is to form an analysis team involving agencies
and stakeholders with sectoral expertise and knowledge of the local context. The analysis team will support joint
analysis with an agreement on the findings and output of the analysis.

The SDR has two main stages, which are broken down further into sub-stages:
• Gathering the Data
✓ Identifying relevant sources
✓ Reading through the sources and extracting information
✓ Storing and organizing this information per topic

15 15
 • Analyzing the Data
✓ Trimming the text by checking the information for reliability and credibility, selecting the most
important information, and the most relevant one for the HBDA
✓ Analyzing the information to identify cross-sectoral priorities, areas and sectors to focus for the
HBDA, especially looking at baseline and preparedness information relating to Shelter
✓ Identifying information gaps, non-assessed areas and sectors
✓ Developing scenarios on the crisis’ evolution, assessing the humanitarian caseload, finding
cross-sector priorities, identifying main assessment needs for the HBDA, including in terms of
coverage and sectors

An SDR is usually structured to present: (1) comparison of pre and post-disaster situations, (2) evaluation of the
quality of the information available, (3) sectoral sheets (relevant sectors should be selected based on the scope of the
HBDA), (4) humanitarian and displacement profiles as relevant, (5) lessons learned from previous similar disasters, (6)
country profile, (7) stakeholder profiles, (8) timeline of events, and (9) discussion of operational and access constraints.
Once information gaps have been identified in the SDR, discussion can begin on what primary data should be collected
in the HBDA, how, and by whom. In the HBDA, it would likely come down to deciding how the data collection tool needs
to be adjusted, in particular with regard to including or not Section B on sectoral needs, and what sampling strategy
needs to be selected (see Part 4.C and see for reference on how to conduct an SDR, ACAPS, 2014).

C. Type of Information and Sources for the HBDA

For the HBDA, the SDR is particularly important as it not only helps collect contextual information on the
disaster or conflict, but also on the pre-crisis situation of residential building and infrastructure. This may
include updated lists of public infrastructure and building locations (schools, hospitals, etc.), as well as on their status
and level of maintenance before the crisis. The goal is to assess if buildings were already showing damage before the
crisis, as well as understand the most common construction materials used in this context, their resilience to the crisis,
the impact of previous disasters on the infrastructure and buildings, and the potential type of disaster waste that may
be present. The SDR also collects information on any initial damage assessments already conducted through remote
sensing.
Common sources of secondary information include other humanitarian and development actors, as well as
government, civil society organizations, and educational and research institutes. Tables 5 & 6 below provide a
list of sources of information for pre-crisis and in-crisis data and GIS. In all cases, it is always important to assess
the secondary sources’ reliability, credibility, and confidentiality (information can be used publicly) before
integrating them into the SDR for the HBDA.
Online resources for SDR and SDR methodology include ACAPS.org, DEEP (deephelp.zendesk.com/hc/en-us),
Humanitarianresponse.info, Humanitarian Data Exchange (HDX, data.humdata.org), Reachresourcecentre.info,
Reliefweb.int, and various cluster-specific websites.

16 16
Table 5: Sources of Information for the Secondary Data Review (Qualitative and quantitative)

Pre-Crisis Sources of Information In-Crisis Sources of Information

Population-related information (demographics and needs) Population-related information (demographics and needs)
o National institutions (ministries, academia, national o Media reports, social media
statistics bureau) o National institutions, ministries, Local Emergency
o Large surveys (DHS, MICS, census) Management Agency (LEMA), chamber of commerce
o International development institutions o Assessment reports from local/ international NGOs,
o National or municipal contingency plans including MIRA, DTMs,
o UN, local/ international NGOs reports and datasets o Funding Appeals, Strategic Response Plans
o Previous Flash appeals, HNOs/HRPs o Situation reports (OCHA, clusters, government)
o Assessments conducted before the disaster by REACH o Assessments after the disaster by REACH Initiative
Initiative available at reachresourcecentre.info available at reachresourcecentre.info
o Analysis, economic or intelligence reports (e.g. o Humanitarian profiles, 3Ws/4Ws/5Ws
cia.gov/library/publications/the-world-factbook) o EU Civil Protection Unit
Building and infrastructure-related information Building and infrastructure-related information
o National or municipal housing bureaus o National institutions, ministries, Local Emergency
o Shelter assessments conducted before the disaster by Management Agency (LEMA), chamber of commerce
REACH Initiative (reachresourcecentre.info) or other o UNDSS, logistic cluster
agencies o EU Civil Protection Unit
Crisis-related information Crisis-related information
o Online databases, the international disasters database o Global Disaster Alerting Coordination Systems/ Virtual
(emdat.be), preventionweb.net, Centre for Research on On-Site Operations Coordination Centre (GDACs.org)
the Epidemiology of Disasters (cred.be) o Copernicus EMS (copernicus.eu/en/services/emergency)
o After Action Reviews o Media reports, social media
o DevInfo, world development indicators, MDGs o Crisis mappers (crisismapping.ning.com), Standby Task
o Monitoring, Early warning and Surveillance systems, e.g. Force (standbytaskforce.org), Qatar computing Research
on unisdr.org, wmo.int, fewsnet.org Institute (QCRI)

Table 6: Sources of Information for the Secondary Data Review (GIS)

Pre-Crisis Sources of Geodata In-Crisis Sources of Geodata

o Common operational datasets (COD), which are primarily
composed of administrative boundaries and populated o Humanitarian OpenStreetMap Team (HOT) tasking
places, available through the Humanitarian Data (updates to OpenStreetMap data)
Exchange (HDX). Note: In some contexts, the national o Event related geodata from the Global Disaster Alert and
authorities will be the best source for this information. Coordination System (GDACS), often courtesy of the
o Infrastructure data from Open Street Map (OSM) JRC, including storm tracks and shake maps.
including buildings, roads, and places of interest (e.g. o Post-event or in-crisis satellite imagery from Google
hospitals, schools, shops, etc.) Note: In some contexts, Earth, ESRI, Maxar, Airbus, Planet or others.
the national authorities will be the best source for this o Post event or in-crisis aerial imagery from OpenArialMap
information. o Imagery analysis from Copernicus EMS, UNOSAT or
o Gridded population datasets from WorldPop, the High- private providers (e.g. Orbital Insights)
Resolution Settlement Layer (HRSL) from o Geodata related to crisis-related data collection exercises
Facebook/Columbia University or the Global Human conducted by other actors (EU Civil Protection, UNDAC,
Settlement Layer (GHSL) from the European national authorities)
Commission Joint Research Centre (JRC).
o Exposure datasets from UNISDR GAR19

17 17
o Pre-crisis satellite imagery from Google Earth, ESRI,
Maxar, Airbus, Planet or others.
o Pre-crisis aerial imagery from OpenArialMap
Key Takeaways
• Conduct a Secondary Data Review (SDR) to establish what is known and unknown about the situation and
its impact and ensure that available data is factored in the HBDA design, the methodology note and data
analysis plan (see chapter 4)
• Specifically collect information on the magnitude of the event – notably for sudden-onset disaster – and
identify what estimations on the impact are already available through satellite imagery
• Note that there are two types of SDR information: pre-crisis (baseline) and in-crisis (already conducted
assessments and information reports about the crisis) – look at Tables 4 & 5 for sources of information
• Based on the SDR, identify information gaps and keep them in mind when selecting the questions for
the data collection tool from the library of questions

18 18
5. Methodology & Assessment Design

A. Methodology and Data Analysis Plan

Each HBDA should have (1) a Methodology Note, (2) a Data Analysis Plan (DAP), and (3) a Data Management Plan
that outlines how data will be handled and by whom during and after the assessment (see Annex 5 for templates). The
Methodology Note and the Data Analysis Plan should also be part of the ToR for the HBDA.
The Methodology Note provides a roadmap for the HBDA, explaining the context and rationale as well as
specifying the final outputs and how they will be achieved. It informs decision-making around methodology and
outputs ensuring a common understanding amongst internal and external stakeholders. It also provides a record of the
agreed objectives of the research, the rationale of the research, the methodology intended to implement it, and the
rationale for choosing the methodology, in particular with regard to the sampling strategy.

A standard set of research questions for the HBDA would include:

o What level of damage on buildings and infrastructure have resulted from the crisis?
o What have been the most affected areas in terms of damage?
o What has been the impact on service provision, e.g. water and electricity, in the affected areas?
o Are there severe needs in Shelter, WASH, Food Security, Nutrition, Education, Protection and Health that warrant
an immediate response and/ or a sector specific assessment?
o What are the prospects for early recovery? In particular, what construction materials are prevalent in the destroyed
buildings?
o What key indicators may inform a PDNA, a Recovery and Peacebuilding Assessment (RPBA), and other similar
assessments?

These questions speak to the main indicators in the HBDA. At the same time, Section B – Building Use and Household
Information would have more granular questions and often information disaggregated by age and gender that would
be needed to be analyzed differently as compared to sections C-G. The strands of information from the different parts
of the HBDA would also need to be cross-analyzed.
The objective of this DAP is to ensure that all research questions are sufficiently addressed through indicators and that
the data collection tool questions/ responses are adequately designed to enable measurement of indicators.
Additionally, the DAP serves as a record of which research questions were supposed to be addressed by which
indicators, thereby helping to focus the interpretation of results during the analysis stage. It is recommended to have a
DAP at least for the main indicators in the HBDA.
The HBDA’s library of questions is structured and is meant to be administered at household level. Table 7
below provides an example of a DAP for the level of damage indicators in the HBDA, showing how they link to
other indicators that can be used to aggregate/ disaggregate the data on the level of damage.

19 19
Table 7: Example of Data Analysis Plan for Selected Indicators for Most Disasters

Data
Research No. Question in Data Response in Data Collection Collection
Indicator
questions # Collection Tool Tool Level/
Method
What is the level
of damages of
residential No damage
E5,
buildings? Have the walls, roof, Minor damage (0-24%)
E6, Household
Estimate level ceiling, floor, foundations, Moderate damage (25-49%)
E8, (HH), Direct
What is the level of damages of the building suffered Severe damage (50-74%)
E9, Observation
of damages of any damage? Completely destroyed (>75%)
E11
different Not possible to assess
components of
buildings?
Where is the building
located? (Country, List of countries, regions, HH, Direct
A1-5 Location
Region, State, City, states, cities, neighborhoods Observation
Neighborhood)
Residential/ Dwelling
Commercial/ Business
Residential & Commercial
Hospital/ Health Clinic
What is the type of the HH, Direct
B1 Type of building School/ Education Facility
Have some building? Observation
Police/ Fire Department
buildings been
Religious/ Faith
more affected
Town hall/ Municipality
than others?
Other
Brick
What factors
Timber
may explain that
Stone
some buildings
D1, Tin/ Zinc sheeting
have been more
D3, What material is the Concrete (column/beam)
affected than
D5, Building building frame, exterior Steel (column/beam) HH, Direct
others?
D13, materials walls, roof, ceiling finish, Concrete plates Observation
D15, floor, foundation? Concrete blocks
(Disaggregation/
D19 Unbaked brick, adobe
Aggregation
Wood, logs
Variables)
Flattened tin cans
Etc.
What are the building’s
C1- HH, Direct
Building Size width, length, height, Enter Number
4 Observation
number of stories?
Detached house/stand alone
Disposition What is the building's
Adjoining house on one side HH, Direct
C6 compared to proximity to other
Adjoining house on two sides Observation
other buildings structures?
Adjoining houses on three sides

20 20
B. Sampling Strategy
Sampling is the process of selecting a number of elements from a larger, defined group. In most assessments,
a sample of population or sites needs to be created because time, resources, and other constraints make it
impossible to assess all populations and sites. The sample selection is informed by the SDR based on resources,
urgency and type of information needs, access, and various other constraints including concern with assessment
fatigue among affected populations and a cost-benefit calculation between the need for information and the potential
protection risks for assessment teams and affected populations. An alternative to sampling is a census that covers
the entire population and/or in the case of a damage assessment all the buildings in the assessed areas (see
for reference on methodology and sampling, CIEDRS, 2003; UNHCR, 2017; and see Annex 1 for definitions).
The HBDA may be implemented using different approaches i.e. census, stratified random sampling, or a
combination between purposive sampling (non-probability method where objects are selected based on pre-defined
criteria, see for reference ACAPS, 2011), snowballing (respondent-driven sampling), and stratified-random
sampling. That choice depends on the objectives of the HBDA, the availability of resources and the type of targeted
buildings (infrastructure vs residential).

It should be noted that the AWG may decide to have a dual approach to the HBDA. This would involve firstly
initiating an assessment during the rapid phase, with a shorter data collection tool using a sample of the population
(e.g. stratified sampling) in order to provide rapid actionable information for the humanitarian response in shelter and
inform other tools such as the PDNA. This would be followed by launching a full-scale more thorough assessment later
during the in-depth phase, which would cover all buildings in the assessed areas (census) in order to inform safe
returns, reconstruction and recovery (see Part 1.C). The approach selected may affect the sampling strategy used, as
a longer survey will require more resources to carry out.

For the HBDA, four scenarios were developed based on resources (personnel, expertise, funding, equipment,
etc.) available to the AWG as well as objective and timing of the assessment, assuming access to the affected
areas and populations. Note that the first scenario is also broken down to allow for the possibility of census as
explained above.
It is ultimately up to the AWG’s judgment call to decide which resources are considered enough for the different
scenarios depending on the scale of the disaster and the qualifications of its’ staff, especially enumerators.

In the first three scenarios, the rationale is the same but the granularity changes. Hence, the scenarios positing
(1) maximum – initial phase only, (2) high, and (3) medium resources assume the use of stratified random
sampling to select buildings to assess across affected areas. As explained above, scenario 1 maximum resources
– in-depth phase is a special case that assumes that instead of relying on a sample, the HBDA covers all buildings in
the assessed affected areas. In scenario 4, the areas to assess are selected using purposive sampling, and only
within these most affected areas, stratified random sampling is used to select buildings.

Table 8 summarizes the scenarios and presents an additional disaggregation between residential buildings and
infrastructure. If there is no access, the HBDA cannot be administered.

21 21
Table 8: Scenarios for Sampling Depending on Resources Available

Infrastructure (schools, hospitals, police stations,

Residential Buildings/ Small Businesses
etc.)
Initial Phase
o Coverage across all affected areas
o Coverage across all affected areas
o Census of all infrastructure
o Stratified random sampling at admin. level 3
o Three options by level of preference:
o Additional stratification by setting (camp-like,
urban, rural) • Based on an accurate, up-to-date list
1. Maximum
o Random GPS points are generated on a map • Based on OSM, Google, other census (see
Resources
and weighted based on population density UNICEF AI project)
(personnel,
o Nearest building unit to each point is targeted • Based on snowballing until saturation
training,
funding, for the survey
equipment) In-Depth Phase
o On the authorities’ request and depending on
information needs (e.g. informing return to o As above
safe buildings), census for residential buildings
and small businesses across affected areas
2. High o Stratified random sampling as above, but no
o As above
Resources stratification by type of settings
3. Medium o Stratified random sampling as above, but o As above, but key infrastructure only, e.g. health
Resources stratification at administrative level 2 only centers and schools
o First (purposive) sampling to select most
o First (purposive) sampling to select most affected areas (based on remote sensing / sat
affected areas (based on remote sensing/ sat analysis, wind speed, distance from epicenter)
4. Limited analysis, based on wind speed, distance from o Coverage across most affected areas only
Resources epicenter, etc.) o Stratified random sampling of key infrastructure
o Coverage across most affected areas only only based on an up-to-date list if available
o Then stratified random sampling as above o If no list is available, snowball sampling until
target is met (i.e. no saturation)
o HBDA does not work in those circumstances o HBDA does not work in those circumstances
o Exploit secondary data, including remote o Exploit secondary data, including remote
No access sensing/ sat analysis, wind speed, etc. sensing / sat analysis, wind speed, etc.
o Support other assessments o Support other assessments
o Execute HBDA once access is restored o Execute HBDA once access is restored

The following section discusses scenarios that require stratified random sampling. When stratified random
sampling is used, the findings can be extrapolated from the study population to a broader population (see
Annex 1 for definitions). Thus, the population is split into non-overlapping groups or strata, and then randomly sampled
within each strata. The purpose is to ensure adequate representation of subjects in each stratum. The strata may be
admin units (sub-district (administrative level 3), district (administrative level 2), cities), settings (camp, urban, rural),
population groups (displaced, non-displaced), building typologies, etc.
By using stratified random sampling, the information gathered from the small number of units visited can be generalized
to the entire population. The more granular the stratification, the better generalizations can be obtained. The sample is
drawn from the sampling frame, a master list of the population or elements to be assessed. Formulas establish a
statistically significant sample size.

22 22
A calculator for sample size is available here (http://www.surveysystem.com/sscalc.htm). Note that when defining a
sample size, it is important to be mindful of the confidence interval (margin of error) and confidence level (see
definitions).

Scenarios 1-3
o Stratification of Residential Buildings: Scenario 1 allows for the most granular coverage by stratifying the
population by both setting and density at the lowest possible administrative unit. Scenario 2 does the same
but without the added stratification by setting. Scenario 3 gives a less precise picture as it stratifies by population
density only at the second administrative level. If resources are available, it is recommended to go with
scenario 1.
o In the case of the HBDA, the selection of points would normally be done by generating random GPS points
on a map covering the population of interest. The distribution of GPS points would be weighted based on
population density should it vary across the targeted area. The building located nearest to each point (within a pre-
defined buffer as relevant to context) is then targeted for the HBDA. If this building cannot for some reason be
assessed, the nearest building to that building is targeted for the HBDA (see box below).
o Assessing Infrastructure: In scenarios 1-2, the objective is to assess all available infrastructure relying on the
list collected through the SDR, and/ or through partners, and/ or generated from asking assessed populations. In
Scenario 3, the objective is to assess only key infrastructure relying on the same methods to identify them. If
resources are available, it is recommended to go with scenarios 1-2.

Scenario 4
o Scenario 4 presupposes that the AWG has severely constrained resources. It relies on non-probability
sampling to select the areas to assess. Since purposive sampling is used, the findings cannot be extrapolated
statistically out of these areas. These areas are selected because they are considered most vulnerable based on
the already known or expected through the SDR analysis on severity of needs, pre-existing vulnerabilities, security
and access, gender, and ethnic diversity.
o Stratification of Residential Buildings: Once the areas for the HBDA are identified through purposive sampling,
the same strategy of stratified random sampling is applied. This would make the findings generalizable across
these areas.
o Assessing Infrastructure: Once the areas of the HBDA are identified through purposive sampling, the key
infrastructure to assess is selected using stratified random sampling as explained above. If no list is available, the
infrastructure is selected by relying on respondents’ responses (snowballing).

No Access
o As noted above, the HDBA cannot be implemented if there is no access to affected areas.

23 23
Note: Assessing Multi-Household Buildings
If section B (Building Use and Household Information) is deployed as part of the HBDA’s data collection tool,
a further on-site randomization may be needed to select households to assess in multi-apartment or multi-
household buildings. This would be done using systematic sampling, given that the sampling frame is known and a
sample size may be agreed in advance. In a ten household or ten apartment residential building, the objective for the
enumerator may be to randomly assess two households (sample of 2). To do so, the enumerator would assess every
10/2= 5th household in the building. To select the first household to assess, the enumerator would use a random number
table or random number generator. The enumerator would then proceed to assess the 5th household in order after the
first household assessed.
Note: GIS–Based Sampling Techniques
Depending on the availability of geospatial secondary data, there are multiple options for utilizing GIS to generate
predefined sample locations based on desired parameters. For instance, it is possible to make either a random
selection of existing spatial data if a building or address registry exists, or generate a new set of random points, in lieu
of having any list to sample from. There are functions in mainstream desktop GIS platforms, including ArcGIS and
Qgis, which facilitate this. Additionally, off the shelf plugins have been developed to make this even easier, notably the
Sampling Design Tool for ArcGIS developed by NOAA, which is available here.
GIS-based sampling offers considerable operational benefits through the predetermination of sample locations and
removal of potential uncertainties that could present significant logistical constraints, including unnecessarily long
distances traveled and time spent on data collection.

C. Adaptation of the Data Collection Tool

The library of questions may need to be adapted in different assessments. There are two main aspects that can
be adapted in the library of questions: the number of indicators and the context-specific indicators.
During that process, the AWG would typically review the information needs; the sectors to be prioritized in the HBDA’s
Section B (Building Use and Household Information); the extent of the geographic areas and populations to be covered;
the capacity and resources of the AWG and of the local authorities; the type of decision-making the HBDA should be
informing; and the type of operational context and disaster (see Parts 2.A, B & 3).
Operational Context
These following sections may need to be especially adjusted based on the operational context.
o Section B: Indicator B52 may need to be adapted to reflect other types of sectors of employment.
o Section D: Indicators D1 to D20 may need to be adapted to reflect other types of building material for buildings in
other contexts.
o Section F: Indicator F22 may need to be adapted to reflect other debris/ disaster waste that could be present in
other contexts; indicator F29 may need to be adapted to reflect other type of hazards common for other contexts.
o Section G: Indicators G1 to G8 may not be relevant for every context and hence may need to be simply removed.
o Across the data collection tool, the assessment team needs to ensure that a common measurement system is
used: either the metric system (i.e. meters, kilometers) or the US customary system (i.e. feet, inches). Depending
on the context, this means adapting the measuring tools and other equipment distributed to the enumerators (see
Part 2.C).

24 24
Depending on the type of disaster, the AWG may discuss the possibility of extensively reviewing the library of questions
to introduce disaster-specific questions for earthquakes, flash floods, etc. As the HBDA toolkit is refined, an additional
library of such questions will be developed.

Number of Indicators
The full library of questions has 243 questions. If all questions are included in the HBDA questionnaire, it will
lead to considerable time spent by enumerators with each building and household. In most cases, it may be
relevant to select certain questions from the library to save time and resources, based on information priorities. This
means removing the indicators that do not cover priority information needs when running a lighter assessment. Across
the library of questions, the priority rating from 1 to 3 for each question gives an indication of the indicator’s importance.
This is nonetheless indicative and may vary based on context and information needs.
Teams interested in running a lighter assessment may typically consider removing parts of or the entire
Section B (Building Use and Household Information), which is the longest part of the library of questions.
Section B is especially time-consuming because it is not observation driven and needs the enumerator to question
individual respondents to collect the information. Additionally, removing parts of or the entire Section G on
preparedness may be another relatively easy way to decrease the length of the questionnaire (see Part 1.B).
As explained in Parts 1.C & 4.B, the AWG may also decide to explore the possibility of having a lighter HBDA rapidly
after the disaster and a more thorough assessment some time afterward. The two assessments would then work in
conjunction to inform different priorities as well as track the evolution of the situation in the affected areas.

Key Takeaways
• The HBDA Core Team produces and the AWG validates a Methodology Note, a Data Analysis Plan, and a Data
Management Plan.
• Key overall questions for the HBDA include: What level of damage to buildings and infrastructure have
resulted from the crisis? What have been the most affected areas in terms of damage? What has been the
impact on service provision in the affected areas? Are there severe needs in WASH, Education, Food Security,
and Health that warrant an immediate response and/ or a sector specific assessment? What are the prospects
for early recovery? In particular, what construction materials were the most used in the destroyed buildings?
What key indicators may inform a PDNA, a RPBA, and other similar assessments?
• Depending on critical data to be collected and available resources, the sampling strategy for the HBDA
changes, leading to changes in the techniques used to select areas to assess. In the majority of cases,
the HBDA relies on stratified random sampling in the areas where it is administered. The HBDA cannot be
administered if there is no access.
• The HBDA’s library of questions should be adapted based on (1) operational context to reflect differences
in construction materials, type of infrastructure, etc.; (2) available resources, timeliness, and information needs
by reducing or increasing the number of indicators used. Note that it is not advisable to use all 240 questions
from the library of questions, due to the time that will be needed to conduct the surveys as well as analyze and
effectively use the results. This will likely only be relevant in exceptional cases.

25 25
6. Data Collection

A. Preparation for Data Collection & Enumerators’ Training

The preparation for data collection includes the HBDA Core Team completing any outstanding HR, administrative,
logistics and related tasks. The HBDA Core Team has also already allocated Assessment Officer(s) to review the early
data coming from the HBDA.
The final preparations to start data collection should also include the HBDA Core Team conducting a final and
detailed review of the access situation specifically in the sites that were selected for assessment. At this
moment, as during the entire process, it is important to remain in liaison with local stakeholders, including authorities if
relevant, NGOs and civil society partners, and community leaders from affected populations.
It is important to have as clear as possible an idea of the day-to-day access constraints in these areas, especially with
regard to security and physical access constraints that can be very dynamic. The enumerators must be made aware
of these constraints, and written consent needs to be obtained from them before they are deployed to these areas (see
for reference on access, ACAPS, 2019).The final and most important stage in preparation for data collection is to
conduct the training of enumerators. The details of using mobile data collection are addressed in Part 3.C. Enumerators
must also to be made aware of general principles of how to collect data in humanitarian assessments, including (see
on accountability to affected populations, Inter-Agency Standing Committee, 2015):
o Confidentiality: Enumerators collecting personal data should handle it with care and not share outside of selected
people in the AWG unless written consent is obtained from the respondents. Personal information can never be
transferred for purposes other than those for which it was collected and for which consent was given. Enumerators
must also be sensitive as to who may overhear interviews.
o Sensitivity: Enumerators should be mindful of potential trauma experienced by respondents, an especially
important aspect in household damage assessments.
o Integrity: Enumerators must treat respondents with decency and respect. They should introduce themselves
respectfully to respondents and be transparent about the goals of the HBDA, its possible outcomes, and how the
data may be used.
o Informed Consent: Enumerators should ask respondents about their consent to use any information they provide.
As part of the HBDA, the enumerators would not be interviewing children. Respondents can always decline to
answer specific question.
o Safeguarding Information: The data and information management activities must adhere to international
standards of data protection. Enumerators should be mindful to not collect more information than is needed for
analytical purposes.
o Participation/ Inclusion: The HBDA Core Team should explain to enumerators the importance of the HBDA
capturing diversity and how this has been incorporated into the sampling strategy if relevant, including with regard
to women, men, girls, boys, persons with disabilities, older persons, and LGBTQ persons.
Besides these principles, enumerators must be explained their role in the HBDA. The length of the enumerator
training depends on the number of enumerators and the length of the data collection tool. In general, it will
vary from one to three days. The HBDA Core Team needs to budget for the logistics of organizing such a training,
including for venue, security, meals, trainers and enumerator salaries. Equipment should be provided and prepared
prior to the training, for example ensuring sufficient numbers of phones and tablets with the appropriate software are

26 26
available, as well as power banks if relevant.As the HBDA toolkit is refined, training modules will be developed on these
aspects.
During the training the enumerators, must be at minimum clear about (see Table 2 in Part 2.C):
✓ HBDA’s context, objectives, time frame, geographical scope, methodology and sampling
✓ Team structure, responsibilities, and reporting lines. This includes assigning team leaders among
enumerators
✓ Techniques and tools to be used, especially the data collection tool and mobile data collection
✓ Techniques on how to implement an ethical and protection-focused approach, including age, gender, and
diversity awareness, including signing a code of conduct and a confidentiality agreement (see principles
above in this section)
✓ Instructions on referral mechanisms for any persons identified as needing urgent attention (e.g.
unaccompanied children, medical cases, survivors of sexual and gender-based violence)
✓ Strategies to enhance data quality and reduce bias when estimating damages, color coding, asking
questions from the preparedness section (G) of the data collection tool
✓ Security protocols and emergency procedures for enumerators’ safety
✓ Administrative and logistics arrangements, such as transport and accommodation

Note: Data Protection and Confidentiality

The Kobo Toolbox team employs industry standard best practices (both technical and administrative) to protect
against unauthorized access of data. Personal information is never shared or sold to third parties. However,
registered users of Kobo are the data controllers of the data they collect using Kobo Toolbox and are responsible
for the safe management of personal information, including compliance with the General Data Protection Regulation
(GDPR). Kobo Toolbox allows users to share application data publicly or with selected users. Information shared
publicly is visible to anyone and can be indexed by search engines. Kobo Toolbox is not responsible for how
registered users handle survey participants’ personal information (see for reference, Humanitarian Response,
2018). If preferred, it is also possible to install an instance of Kobo Toolbox on a privately owned and managed
server. For example, UNHCR currently have their own server, and some governments may prefer data to be stored
on their own servers.

B. Day-to-Day Management
The HBDA is managed on a day-to-day basis as any other humanitarian assessment/ project (see for reference on
assessment management, UNHCR, 2017). The best practices in the field include:
✓ Keeping track of the data collection process, including in terms of areas coverage and number of surveys
completed, with AWG staying in contact with enumerators in the field through their team leaders if such
were assigned
✓ Monitoring the local context to be aware of any changes in security and access conditions
✓ Keeping in daily contact with local counterparts among local NGOs, local and national authorities

27 27
 ✓ Keeping an updated registry of progress and encountered challenges, including logistics, costs (e.g. salary,
per diem, procurement of equipment, vehicles, etc.)
✓ Starting cleaning and processing the early collected data to catch mistakes, misunderstandings, lapses in
coherence/ debriefing part of the enumerators to be sure indicators are understood by everyone
✓ If issues are identified, update the data collection tool and re-train enumerators or circulate guidelines

C. Mobile Data Collection and Kobo Toolbox

Data in the HBDA is collected in the field through an online questionnaire using Kobo Toolbox (an open source
data collection platform on electronic devices that allows rapid information gathering and facilitates the post-collection
data analysis). The data collected is then later visualized in Microsoft Power BI (an intelligence platform used for the
analysis and visualization of data). Other information management and visualization tools, including Microsoft Excel,
Tableau, ArcGIS, and Adobe Illustrator may also typically be used (see Part 6.C and Annex 3).
Assessment and GIS Officers must be proficient in the use of the Kobo Toolbox in order to train enumerators on mobile
data collection. During the training, enumerators must be shown how the Kobo Toolbox and mobile data collection
work in the Kobo application as well as the phone’s internet browser. The following link (XX) gives access to the HBDA
data collection tool.
During the training, enumerators should be made aware of the limitations associated with the HBDA, in particular
regarding GPS localization during mobile data collection using the Kobo Toolbox (see Part 1.D). Two additional points
warrant special attention:
✓ Enumerators need to know that once the data collection tool is open, no internet connection is needed to
register the information. However, for better GPS accuracy, it is recommended to always have internet
access. If there is no internet access, the information collected is stored on the phone and uploaded once
there is access to internet. To upload the information, the phone needs to be connected to the internet with
the link open for at least 10-15 min. Time varies depending on internet speed and size of the pictures.
✓ The HBDA data collection tool in the Kobo Toolbox uses a color coding for damage which it assigns based
on the information entered. The enumerator is asked to confirm the suggested color tagging such as: Green:
Minimal damage; Yellow: Medium damage; Orange: Major damage, significant repairs required; and Red:
Destroyed structure, probably requiring complete rebuilding. It is important that if the enumerator disagrees
with the tagging, they return back in the data collection tool and change the information for damage questions
instead of simply overriding the color coding suggested.
As the HBDA toolkit is refined, a specific guideline will be developed for the training of enumerators in the HBDA.

Key Takeaways
• The training of enumerators ensures the accuracy of the HBDA. Enumerators have to be briefed on:
o General principles (confidentiality, sensitivity, integrity, informed consent, safeguarding information,
inclusion) on how to collect data in humanitarian assessments
o Practical aspects of the HBDA, e.g. objectives, security, techniques, the data collection tool, color
coding, etc.

28 28
 o The use of the Kobo Toolbox in mobile data collection
• The HBDA Core Team, part of the AWG, handles the day-to-day management of the HBDA, including
keeping track of data collection, checking the quality of the early data, monitoring the context for changes in
access, keeping in touch with local authorities and other stakeholders.

7. Data Analysis

A. Data Cleaning and Processing

The AWG set procedures for collation, quality control, and processing of data. It establishes protocols among partners
for the sharing of different types of information. The protocols on data cleaning and processing need to at minimum
define:
✓ Type of metadata to describe the dataset (date of collection, geographic coverage, methodology)
✓ Geographical units (use of country P-codes and/or agreed administrative place names, disaggregation levels,
and other technical standards)
✓ Ownership the data, staff’s rights to change and/ or modify the data
✓ Determination of whether datasets should be cleared or anonymized prior to sharing, and by whom
✓ Rights to access the data at each level of aggregation and sensitivity
✓ Details on data confidentiality and safeguarding of information

After these protocols are agreed and implemented, members of the HBDA Core Team, usually Assessment
and Information Management Officers, clean the data. The objectives of data cleaning are multiple, including:
✓ Checking for data consistency, including that there are no significant gaps or logical inconsistencies in the data,
and checking the use of the “other” and “specify” options across the data collection tool
✓ Checking for data coherence and plausibility, e.g. comparing collected data to baseline pre-crisis information,
and/ or findings from similar contexts especially for harmonized sectoral indicators, and/ or existing humanitarian
standards
✓ Pre-identifying the most useful indicators and ensuring that data for these indicators is present
✓ Ensuring that data is geographically tagged, including that GPS readings are coherent
✓ Ensuring that data collection standards were respected, including metadata, names of enumerators, dates, and
unique identifying numbers are present
✓ Ensuring that methodology and sampling choice were respected
✓ Anonymizing the data by removing personal information on respondents if need be before sharing

As a rule, the Assessment and Information Management/GIS Officer(s) in charge of cleaning the data will maintain a
data cleaning log, and consult with the Needs Assessment Coordinator before altering the data. The data cleaning

29 29
needs to be done daily, with a full comprehensive final review of the whole dataset after data collection has finished
(see for reference on data cleaning, ACAPS, 2016 and Annex 5 for templates).

B. Data Analysis
The analysis part is one of the most challenging and crucial parts of any HBDA. The quality and the relevance of the
findings depend on the quality of the data collected but also on the ability of the Assessment and Information
Management Officer(s) conducting the data analysis.
The analysis aims to make sense of the information collected and to draw conclusions about the severity of
conditions and humanitarian needs priorities. There are four different phases of analysis (see for reference on
analysis, UNHCR, 2017):

o Description: Grouping, summarizing, and comparing data to identify trends, patterns, and outliers. Comparison
of the following type are most commonly used in humanitarian data analysis to describe the data:
➢ To existing common conventions and/ or standards (e.g. SPHERE minimum standards, pre-crisis
standards, standards in neighboring countries with similar conditions)
➢ Across geographical units (provinces, districts, sub-districts, neighborhoods, villages), areas of high/
limited conflict-intensity, urban/ rural areas, camps and other settings
➢ Across socio-economic and ethnic/ national groups (e.g. agro-pastoralists vs. farmers, hosts vs.
IDPs, ethnic group A vs. ethnic group B)
➢ Based on sex and age, and other potential factors of marginalization and accrued need, including
persons with disabilities, LGBTQ, minorities, and indigenous people
➢ Based on evolutions in time (from the previous crisis to this one, from pre-crisis to the moment of the
HBDA, from the last HBDA to the current HBDA)
o Explanation: Determining the underlying factors and processes behind certain trends and patterns. The aim is to
identify associations, correlations, causations, and any types of connections in the data of specific indicators and
across indicators. Data analysts may develop different hypotheses to explain the data that need to be discussed
collectively within the AWG afterward.
o Interpretation/ Operationalization: Drawing conclusions on severity and priorities based on available evidence,
and estimating the extent to which the findings can apply to others settings. The aim is also to start preparing
operational recommendations as part of the decision-making based on the HBDA.
➢ Since the HBDA usually relies on stratified random sampling, the HBDA Core Team needs also to identify
the extent to which and how the findings may apply to non-assessed areas and populations. This involves
a review of the methodology and a debriefing of the enumerators to capture qualitative information based
on direct observation.
➢ The team needs to identify remaining information gaps and the limitations of the data.
o Early Warning & Preparedness: Predicting or forecasting the crisis’ likely evolutions based on past and present
data. The aim is to anticipate potential future events and their consequences, and help identify new and emerging
risks. This may include developing scenarios to estimate future the evolutions of the crisis and future humanitarian
caseloads. In the HBDA, this step is also about preparing early recovery and conducting preparedness efforts for
upcoming crises (see for reference on scenarios building, ACAPS, 2013).

30 30
The interpretation and early warning phases are often conducted jointly by the AWG, possibly as part of a workshop
with additional key stakeholders. Afterward, the main findings of the data analysis are compiled in information products
that are shared, which may include reports and/or factsheets (see Part 6.D). They are the basis for subsequent
response analysis that draws strategic, programmatic, and/ or operational conclusions based on current and forecasted
severity and priorities.
The following section discusses some basic analysis of key indicators in the HBDA across the data collection tool. Most
of them would be presented as percentage of the assessed population.
B – Building Use and Household Information
Check the ratio of residential to non-residential buildings in the HBDA (B1), and evaluate if non-residential buildings are over-
assessed. If they are, propose possible explanations
Household Information – Identify the male/female ratio among respondents (B8); the age distribution of the respondents (B10);
the average household size (B12/13); the number of IDP (B14-16)
Household Health and Safety – Identify people with acute or special needs (B17-36)
Employment, Income, and Food Security Situation of Household – Identify needs in food security, including needs that may
pre-date the crisis (B37-44), and the potential for cash-based programing based on the availability of markets (B43-44). Identify
the impact of the crisis on livelihoods (B45-53) and the support received (B54-57) and the potential to be engaged in work as part
of recovery activities (B58). Calculate variations in food security needs as compared to the pre-crisis baseline; identify areas with
acute needs (see “Description” in Part 6.B)
Household relationship to building – Identify if people may receive insurance to recover from the crisis (B59-62)
Commercial or Mixed Use – Check that no category is over-represented among assessed buildings (B63), and if is, think of
explanations. Estimate percentages of business that stopped operations following the crisis, the impact it had on local
employment, and treat this as an indication of the economic health in the assessed areas (B69-76)
Non-Residential – Check that no category is over-represented among assessed buildings (B77), and if it is, think of
explanations; assess especially if non-residential buildings are being used to host IDP (B91-93)
Education Facility – Check that no category is over-represented among assessed buildings (B94), lack of secondary and higher
education facilities in assessed areas gives information on the socio-economical, situation of the population in that area; assess
needs in education as shown by lack of teachers (B96-97), damages to premises (B98-99), and furniture (B100-103); assess the
catchment population of the education facility (B104-123). Calculate ratios of teacher per student; identify areas with lack of
education coverage (see “Description” in Part 6.B)
Health Facility – Check that no category is over-represented among assessed buildings (B124), lack of hospitals/ health clinics
suggests pre-existing needs in health. Health facilities may lack the needed number of beds (B133-134) and/ or personnel
(B131-132), a situation that may have also pre-dated the crisis. Pre-existing needs may have been exacerbated by the crisis due
to damages (B126-128, B137), and/ or important caseload in health (B136). Calculate ratios of doctors/ nurses per patient;
identify geographical areas with limited health coverage/ most acute needs (see “Description” in Part 6.B)
D – Building Materials
Calculate the materials most needed for repairs for the different parts of the building across all households but also
disaggregating by different parameters (see “Description” in Part 6.B)
E – Building Damage and Repairs
Damage – Identify the ratio of buildings showing signs of damage before the crisis (E1); ratios on the levels of damage and
geographical repartition, and disaggregation by other factors (E5-14) (see “Description” in Part 6.B); assess the situation with
potential flooding in case of certain sudden-onset disaster and ensure the data is consistent across enumerators (E10)
Building Repairs – Evaluate the state of early recovery (E18-24), evaluate especially the ratio of houses unable to do repairs
due to either lack of financial support and/ or materials (E19)
F – Building Services and Debris

31 31
Building Services – Evaluate the situation with power and electricity availability during the day especially in public buildings,
e.g. schools and hospitals (F1-6); evaluate the situation with WASH in households and especially in public buildings (F9-18),
compare to SPHERE standards; evaluate use of unsafe water sources; (see “Description” in Part 6.B) for disaggregation options
Debris Management – Estimate the amount of debris that needs to be collected; estimate the extent to which recovery and
debris/ disaster waste cleaning activities may be challengingt; evaluate the coherence of the information in F21 on amounts of
debris, and disaggregate by geographical area; evaluate the risks for debris cleaning activities (F30) (see for reference on debris
management, UNDP, 2015)
G – Building Preparedness
Information is partly subjective since it is based on respondents’ perception; it needs to be triangulated against SDR information.
C. Analysis Tools for Visualization
Microsoft Power BI
Microsoft Power BI is an industry standard business intelligence platform which, due to its ease of use, offers a lower
entry point for data visualization and analysis. As such, it has been identified as the preferred platform for HBDA
analysis and visualization. Learning materials are readily available for both the web and desktop applications. Data can
be easily sorted, filtered, analyzed, mapped and arranged. The example below presents a dashboard for Hurricane
Dorian in the Bahamas in August 2019.

As the HBDA is rolled-out, Power BI dashboard templates will be adapted to each context and improved upon as
lessons learned are incorporated. The HBDA toolkit will also be refined and specific guidelines on the use of Power BI
will be produced.

Geographic Information Systems (GIS)

While Microsoft Power BI is the preferred platform for analyzing and visualizing HBDA data, conventional GIS platforms,
including ArcGIS or Qgis, can help further analysis and fully leverage the spatial element of the data collected.
Typical analyses aggregate collected data to desired boundaries (e.g. neighborhood), identify linkages to other spatial
datasets, or calculate density of data points creating a “heatmap”. Geographic information products themselves can

32 32
range from stand-alone “static” maps to interactive web maps to embedded maps in other applications, including Power
BI.
While not overly complicated, inclusion of this type of analysis would generally necessitate the incorporation of GIS
staff in the HBDA team (see Annex 4 for the ToR of a GIS Officer). Having trained GIS staff would also be beneficial
for developing standalone or other map templates in advance to facilitate rapid dissemination of GIS analytical products
(see Annex 3 for a list of GIS applications and links to related resources).

D. Information Products, Target Audience, & Dissemination

After analysis is completed, the AWG needs to establish a list of envisioned outputs for the HBDA, from the
final comprehensive report, to factsheets and profiles, dashboards, infographics, maps, the methodology, and
the dataset itself. The process of disseminating the findings is key to translating the HBDA into decision-making in
the humanitarian field. Different products would be targeted at different audiences and may be disseminated in different
ways. Protocols for data sharing and confidentiality may be needed to regulate the sharing of the HBDA’s dataset.
Since data in humanitarian assessments becomes obsolete quickly, it may be worth sharing preliminary
findings with partners and stakeholders before the final report and other information products are released. It
is also important that channels of communication remain open to the assessed populations.

The HBDA comprehensive report is usually drafted by the Assessment Officer(s) under the supervision and
with the review of the Assessment Coordinator. It needs to include the following information and components:
✓ Objectives and scope of the HBDA
✓ Description of the populations and areas covered in a map, and the dates of data collection
✓ Description of the methodologies and sampling strategy, including their limitations
✓ Context description including the scale of the crisis and the socio-political context
✓ Findings of the HBDA and AWG’s conclusions and recommendations
✓ Appropriate visualization showing at least the geographic repartition of buildings based on levels of damages
✓ Indication of needs, priorities, and challenges disaggregated by different factors (see Part 6.B)
✓ Note on gaps and challenges in specific sectors, geographic areas, and population sub-sets based on analysis
of the severity, depth, and estimated prevalence of the problem; and a comparison of current living standards
with a baseline or established benchmarks based on the SDR
✓ Acknowledgement of all the international, national, and local stakeholders who participated in the HBDA

The AWG then decides on the appropriate channels to disseminate the final comprehensive reports and the
other different information products of the HBDA. To facilitate this, the AWG should consider working with country-
level UNDP or other relevant agency Communications Officers to ensure appropriate external communication. It should
be noted that as the HBDA is conducted upon request, data is usually collected on behalf of and for governments.
Governments should be involved in dissemination decisions, and the AWG can assist with identifying relevant
humanitarian partners and dissemination channels. Specific channels for dissemination include:
✓ UNDP colleagues from country office and headquarters

33 33
 ✓ Local and national authorities depending on the context
✓ Humanitarian partners among UN agencies, clusters, local and international NGOs depending on context
✓ Humanitarian partners in headquarters, including people in advocacy and communication
✓ Humanitarian websites such as HDX, Humanitarian Response, Refworld, and Reliefweb
✓ Cluster-specific websites (e.g. sheltercluster.org, globalprotectioncluster.org, globalcccmcluster.org)
✓ Social media such as Facebook, Instagram, Twitter, LinkedIn, Yammer
✓ File synchronization services such as SharePoint, Dropbox, and the Humanitarian Kiosk

Key Takeaways
• Data cleaning is an important step that needs to be conducted before analysis can start. The main objectives
of data cleaning are to check for consistency, plausibility, and appropriate geographical tagging.
• High quality data analysis is one of the most challenging parts of the HBDA. It helps draw conclusions about
the severity of crisis and humanitarian needs priorities. There are four phases of analysis:
o Description – Comparing the HBDA data to other datasets (pre-crisis baseline in SDR, humanitarian
standards, and across geographical units, population groups, and over time)
o Explanation – Postulating hypotheses to explain the data, identifying causes, and connecting
indicators among themselves.
o Interpretation – Drawing conclusions about severity of the crisis, most affected groups, identifying
priorities in affected areas, and preparing operational recommendations.
o Early Warning – Predicting the evolution of the crisis, planning for early recovery, and anticipating
future crises by setting-up early warning and monitoring systems.
• Microsoft Power BI is a unique tool to visualize the data from the HBDA that is easy to use; there are
dashboard templates for the HBDA using Power BI.
• Multiple outputs are possible based on the HBDA, including reports, factsheets, dashboards infographics,
and maps. Different products are targeted at different audiences, taking into account protocols for data sharing
and confidentiality limitations with regard to the data collected during the HBDA.

34 34
Annex 1: Definitions

o Census: While sample surveys result in information obtained from only a subset of a population, a census results
in every person being counted and registered individually (MSF 1997, MSF 2006). A census covers the entire
population of a country; in addition to individual data, a set of relevant socioeconomic information is gathered for
every household (IDMC/OCHA 2008). (ACAPS, 2012; UNHCR, 2017).
o Common Operational Dataset (COD): Critical datasets that are used to support the work of humanitarian actors
across multiple sectors. CODs contain important geographic data like administrative boundaries, populated places
(settlements), transportation networks, and information about water bodies and terrain. Population statistics and
humanitarian profile data are also included. CODs are considered a standard for the humanitarian community and
should represent the best available datasets for each topic. (UNHCR, 2017)
o Confidence Interval (Margin of Error): The plus-or-minus figure that bracket results. For example, if you use a
confidence interval of 4 and 47% percent of your assessed population picks an answer you can be "sure" (See
confidence level) that if you had asked the question of the entire relevant population between 43% (47-4) and 51%
(47+4) would have picked that answer. (Surveysystem.com, 2019)
o Confidence Level: It is the percentage number that tells you how sure you can be in your findings. It represents
how often the true percentage of the population who would pick an answer lies within the confidence interval. The
95% confidence level means you can be 95% certain; the 99% confidence level means you can be 99% certain.
Most humanitarian researchers use the 95% confidence level. (Surveysystem.com, 2019)
o Coordinated Needs Assessment: Assessments that are planned and carried out in partnership with other
humanitarian actors, with the results shared for the benefit of the broader humanitarian community to identify the
needs of the affected population of a humanitarian crisis. Coordinated assessment is a broad term that includes
several different types of assessments, ranging from inter- and intra-cluster/sector joint assessments to single
agency assessment that are harmonized. (IASC, 2012)
o Data Analysis Plan: The tool operationalizing the analytical framework providing sources, methods, and data
collection techniques for each information element sought, as well as how the information will be analyzed.
(UNHCR, 2017)
o Debris: Mixture of building waste and rubble typically arising from damaged buildings and their demolition. This
waste stream can include natural materials such as clay and mud, trees, branches, bushes, etc.
o Disaster Waste: Solid and liquid waste generated from a disaster, not limited to debris. It includes concrete, steel,
wood, clay, tar elements from damaged buildings, infrastructure, household furnishings, parts from power and
telephone grids such as electrical poles, wire, electronic equipment, transformers, parts from water and sewerage
distribution centers, natural debris such as trees, mud and plants, chemicals, dyes and other raw materials from
industries and workshops, waste from relief operations, damaged cars, bicycles, unexploded ordnances,
packaging materials, pesticides, paint, varnish, solvents, and healthcare waste. (UNDP, 2015)
o Disaggregation, includes SADD (Sex and Age Disaggregated Data): Date is separated along specific sets of
variables or criteria (e.g. age, sex or economic status). (UNHCR, 2017)
o Emergency Employment: Job opportunities generated by national governments, international organizations and
NGOs for the affected populations in response to arising humanitarian crises. (ILO, 2019)
o Enumerator: Person in charge of administering the HBDA to the affected populations, and conducting the direct
observation of damages to buildings and infrastructure.

35 35
o Fundamental Operational Dataset (FOD): Datasets that are relevant to a humanitarian operation whose content
is not covered by CODs. FODs often contain data that is specific to a particular sector, such as schools, wells, or
security incidents. (UNHCR, 2017)
o Geographic Information System (GIS): An organized collection of tools (computer hardware and software),
information, and professional/ technical knowledge used to input, store, retrieve, utilize, analyze, and output
geographically referenced data. GIS uses geography as its organizing principle. GIS is useful in situations with a
spatial dimension, such as knowing the locations of refugees and mapping damages to buildings. (UNHCR, 2017)
o Harmonized Needs Assessment: A needs assessment in which data collection processing and analysis is
undertaken separately from other needs assessments, however the data is sufficiently comparable (because of
the use of Common Operational Datasets, key indicators, and geographical and temporal synchronization) to be
compiled into a single database and to serve as the subject of a shared analysis. (IASC, 2012)
o Household: A household is defined as a housekeeping unit or, operationally, as a social unit that has a (1)
common arrangements for food; (2) shares household expenses or daily needs; and (3) lives in a shared dwelling.
A household includes either one person or a group of people, not necessarily related, living at the same address
with common housekeeping i.e. sharing at least one meal per day or sharing a living room.
o Household Membership: In the context of the HBDA, households are classified according to: (1) Main household
members i.e. persons in the dwelling who are its long-term permanent occupants and (2) Temporary household
members: persons who have gotten temporary shelter after being displaced by the crisis. Following on the above,
the unique household ID number relates to the main household members.
o Main Income Earner: The main income earner is the person who contributes the largest portion of household
income, covering most household expenses and financially supporting other household members.
o Main Respondent: The main respondent is an adult in the household who makes decisions about the household
and can be contacted for follow up. The main respondent has knowledge about the household and/ or is the
responsible adult in the household when the enumerator comes to administer the survey. During the HBDA, the
adults in the household may not be the main decision makers. In that case, the most knowledgeable about
household affairs should be interviewed as the main respondent.
o P-Code: Short for ‘place code.’ P-codes provide unique reference codes to geographic locations and are important
identifiers in data management systems. (UNHCR, 2017)
o Purposive Sampling: A non-probabilistic sampling method in which target subjects with different characteristics
are selected to fit a particular purpose. Results based on purposive sampling cannot be generalized to cover an
entire population. (UNHCR, 2017)
o Random Sampling, including stratified and systematic: This is a method to draw a representative sample by
means of selecting households or individuals randomly from the whole population of households or individuals
surveyed. (UNHCR, 2017) (see Stratification)
o Secondary Data Review (SDR) (or Desk Review): Review of information already available. A desk review is
always an advisable first step to understand the information needs in a particular context. (UNHCR, 2017)
o Snowball Sampling: Households and individuals are selected according to recommendations from other
informants, with each informant recommending the next set of informants. (WFP, UNHCR, 2017)
o Stratification: Dividing the assessed population into several sub-groups that share common characteristics, in a
way that each individual only belongs to one sub-group (with no overlap among them). For example, different age
brackets can be used to divide the population into sub-groups of children, teenagers, adults, and elderly. (UNHCR,
2017)

36 36
Annex 2: Reference Documents

ACAPS, Data Cleaning, April 2016.

(https://www.acaps.org/sites/acaps/files/resources/files/acaps_technical_brief_data_cleaning_april_2016_0.pdf).
ACAPS, Humanitarian Access, Methodology Note, September 2019.
(https://www.acaps.org/sites/acaps/files/resources/files/20190910_acaps_access_technical_brief_september_20
19_update_v2.pdf).
ACAPS, Purposive sampling and site selection in Phase 2, October 2011. (http://www.coordinationtoolkit.org/wp-
content/uploads/ACAPS-Purposive-Sampling-and-Site-Selection.pdf).
ACAPS, Scenario Development, Summary Brief, 2013.
(https://www.acaps.org/sites/acaps/files/resources/files/scenario_building-pocket_version.pdf).
ACAPS, Secondary Data Review, Technical Brief, May 2014.
(https://www.acaps.org/sites/acaps/files/resources/files/secondary_data_review-
sudden_onset_natural_disasters_may_2014.pdf).
ACAPS, Survey of Surveys, Technical Note, July 2011.
(https://www.acaps.org/sites/acaps/files/resources/files/technical_note-survey_of_surveys_july_2011.pdf).
Center for International Emergency, Disaster and Refugee Studies, and Hopkins Population Center, Demographic
Methods in Emergency Assessment, A Guide for Practitioners, 2003. (http://www.coordinationtoolkit.org/wp-
content/uploads/CIEDRS-Demographic-Methods-in-Emergency-Assessment.pdf).
Inter-Agency Standing Committee, Multi-Sector Initial Rapid Assessment Guidance, July 2015.
(https://www.humanitarianresponse.info/en/programme-cycle/space/document/multi-sector-initial-rapid-
assessment-guidance-revision-july-2015).
Inter-Agency Standing Committee, Operational Guidance for Coordinated Assessments, March 2012.
(https://www.humanitarianresponse.info/sites/www.humanitarianresponse.info/files/documents/files/ops_guidance
_finalversion2012_1.pdf).
Inter-Agency Standing Committee, Protection and Accountability to Affected Populations in the HRP, 2015.
(http://www.globalprotectioncluster.org/_assets/files/tools_and_guidance/protection_of_civilians/edg-
aap_protection_guidance_note_2016.pdf).
Kobo Toolbox, Help Center, 2019. (http://support.kobotoolbox.org/en/).
UN-Spider, Knowledge Portal, 2019. (http://www.un-spider.org).
UNDP, Guidance Note on Debris Management, November 2015.
(https://www.undp.org/content/undp/en/home/librarypage/crisis-prevention-and-recovery/signature-product--
guidance-note-on-debris-management.html).
UNDP, Post-Disaster Needs Assessment, September 2014.
(http://www.undp.org/content/undp/en/home/librarypage/crisis-prevention-and-recovery/pdna.html).
UNHCR, Needs Assessment Handbook, June 2017. (http://needsassessment.unhcr.org/media/downloads/2017-
05_NAH_BOOK_NEW.pdf).

37 37
Annex 3: Information Management, Visualization, Mapping and Navigation Tools

o Microsoft Excel: An indispensable tool for data/ information management. While not listed in the Analysis tools
section explicitly, this software is generally expected to be part of the HBDA process/ workflow.
o ESRI ArcGIS: ArcGIS is the industry standard desktop GIS application. There are now two versions available,
Desktop and Pro. Both offer similar features that can be used to enhance the HBDA with additional mapping
products. There is also a plug in available for Power BI which enables the incorporation of custom maps. Further
a number of applications have been developed by ESRI that could be useful for HBDAs:
o Survey 123 for ArcGIS: Survey 123 is a custom data collection application, built with integration with
GIS in mind and supports the XLSForm specification.
o Workforce for ArcGIS: Workforce is an application designed for managing enumeration teams and can
more effectively distribute work through the creation of discrete assignments.
o Operations Dashboard for ArcGIS: Operations Dashboard a configurable web app that provides
location-aware data visualization and analytics for a real-time operational view of people, services,
assets, and events.
o QGIS: Quantum GIS or QGIS is a free and open source desktop GIS application. It has a wide following and
functionality has increased in recent years to make it largely comparable with ArcGIS. QGIS is still a powerful tool,
particularly for projects/ missions with smaller budgets. The only downside is limited documentation in comparison
to ArcGIS and some bugs which take longer to resolve as it is an open source application.
o Adobe Illustrator: Adobe Illustrator is a premier vector design program that can facilitate the production of high-
end graphic visualizations.
o Adobe InDesign: InDesign is Adobe’s desktop publishing application with allows for the production of rich graphic
documents.
o ODK Collect: ODK Collect is the original Android application created for OpenDataKit data collection. ODK collect
can be linked to a Kobo Toolbox server and it is possible to customize offline maps for the application.
o ODK Briefcase: ODK Briefcase is an offline data aggregation application, very useful in low or no bandwidth
environment.
o GeoODK: GeoODK is an ODK based data collection application geared towards collection of geospatial
information and integration with GIS applications.
o Open Map Kit: OpenMapKit is an ODK based data collection application built with OpenStreetMap in mind and
offers the possibility to create an OpenStreetMap question which launches the OMK app to enable OpenStreetMap
editing.
o Android Mapping and Navigation Applications: There are many applications, notably Maps.me, SW Maps,
MAPinr, which offer the ability to load a geodata related to data collection (e.g. affected area, daily assignment,
etc.) on offline extracts of resources like OpenStreetMap.

38 38
Annex 4: Terms of References (ToR) of the HBDA Team

ToR, Program Lead/ Assessment Coordinator

ToR, Assessment Officer
ToR, Information Management/ GIS Officer/ Remote Sensing Analyst

39 39
Annex 5: Templates

Data Cleaning Logbook

Question
UID Issue Feedback Changed Old Value New Value
Name

XXXX1 B.1. Enumerator reports Recode to “Commercial/ Yes Other- “Commercial/

Type_Buil "Other- Business” Supermarke Business”
ding Supermarket” t

XXXX2 F.14 Enumerator reports Reviewed with enumerator, Yes 60 6

Functionin 60 input error
g_Latrines

XXXX3 E.6 Enumerator reports Reviewed with enumerator, Yes Completely Severe
Roof_Da over 75% for almost damages is under-estimated destroyed damage (50-
mage all buildings he in that area and enumerator (>75%) 74%) (partly
assessed despite over-estimates damages, (collapsed) collapsed, not
his assessment need to review his answers structurally
area not being one and triangulate with other sound and not
of the most affected enumerators from that area livable)

XXXX4

Data Management Plan

Documentation and Metadata
Data Cleaning Log, including:
What documentation and □ Data analysis plan □ □ Deletion Log
metadata will accompany □ Value Change Log
the data? □ Code book □ Data Dictionary
Select all that apply Metadata based on HDX
□ □ [Other, Specify]
Standards
Ethics and Legal Compliance
□ Consent of participants to participate □ Consent of participants to share personal
information with other agencies
Which ethical and legal No collection of personally identifiable Gender, child protection and other
measures will be taken? □ □
data will take place protection issues are taken into account
All participants reached age of
□ [Other, Specify]
majority
Who will own the
copyright and Intellectual
[Specify]
Property Rights for the
data that is collected?
Storage and Backup

Where will data be □ HBDA/ AWG Kobo Server □ Other Kobo Server: [specify]
stored and backed up
during the research? □ Government Server □ Physical location [specify]

40 40
 □ On devices held by UNDP Staff □ [Other, Specify]
Data access is limited to [Specify,
□ Password protection on devices/
□ e.g. UNDP staff, HBDA Core Team,
servers
Which data access and AWG]
security measures have □ Form and data encryption on Partners signed an MoU if
been taken? □
data collection server accessing raw data

□ [Other, Specify]

Kobo Access Rights

Kobo Access Person Account Name

View Form [Insert name] [Insert account name]

View and Edit Form [Insert name] [Insert account name]

View Form and Submit
[Insert name] [Insert account name]
Data
[Insert account name] [Download data
Download Data [Insert name] access to a deployed form can only be
awarded to one single individual.]
Raw Data Access Rights
Raw Data Access Reason Person

[Insert name] [must be the same person as

Accountable Accountable the one who has “download data” rights in
the Kobo Access Rights table]
[Explain why this person needs to
Access access to raw data, e.g. GIS: [Insert name]
choropleth maps using GPS points]
[Explain why this person needs to
[Add relevant number of rows
access to raw data, e.g. GIS: [Insert name]
for access rights]
choropleth maps using GPS points]
Preservation
Where will data be □ Government Server □ OCHA HDX
stored for long-term
preservation? □ UNDP Country Server □ [Other, Specify]
Data Sharing
Will data be shared
□ Yes □ No
publicly?
Only anonymized/ cleaned/
□ Yes □ consolidated [delete what does not
If yes, will all data be
apply] data will be shared
shared?
□ Other [Specify]

Where will you share the □ Government □ OCHA HDX

data?

41 41
 □ Humanitarian Response □ [Other, Specify]

Data Protection Risk Assessment

No, no information that allows
Have you completed the □ Yes □
identification of individuals is collected
Indicators Risk Assessment
table below? [Complete the first 4 columns in the Indicators Risk Assessment table below]

Type of Disclosure Required

Risk Indicator Benefits Class
Identification Risk Implications Mitigation
[Specify identification [Specify implications, [Specify
[Specify indicator, [To be
risk, e.g. Direct e.g. loss of privacy/ benefits, e.g. [To be specified
e.g. HH_phone completed by
contact/identification potential target of follow up for by AWG]
number] AWG]
of HH] armed actors] authorities]
[Add relevant
number of rows
for risk indicators]
Responsibilities
Data collection [Insert name, function, email]

Data cleaning [Insert name, function, email]

Data analysis [Insert name, function, email]

Data sharing/uploading [Insert name, function, email]

Methodology Note
Country
Type of Emergency □ Natural disaster □ Conflict
Type of Crisis □ Sudden onset □ Slow onset □ Protracted
Research Timeframe 1. Start collect data: _ _/_ _/_ _ _ _ 5. Preliminary presentation: _ _/_ _/_ _ _ _
2. Data collected: _ _/_ _/_ _ _ _ 6. Outputs sent for validation: _ _/_ _/_ _ _ _
3. Data analysed: _ _/_ _/_ _ _ _ 7. Outputs published: _ _/_ _/_ _ _ _
4. Data sent for validation: _ _/_ _/_ _ _ _ 8. Final presentation: _ _/_ _/_ _ _ _
Number of HBDAS □ Single assessment
□ Multi assessment (Rapid + In-Depth, Monitoring)
[Describe here the frequency of the cycle]
Milestone Deadline
□ Donor plan/strategy _ _/_ _/_ _ _ _
Humanitarian
milestones □ Inter-cluster plan/strategy _ _/_ _/_ _ _ _
Specify what will the □ Cluster plan/strategy _ _/_ _/_ _ _ _
HBDA inform and when
□ NGO platform plan/strategy _ _/_ _/_ _ _ _
□ Other (Specify): _ _/_ _/_ _ _ _
Audience type Dissemination

42 42
 □ General Product Mailing (e.g. mail to NGO
consortium; HCT participants; Donors)
Audience Type & □ Strategic
□ Cluster Mailing (Education, Shelter and WASH)
Dissemination Specify □ Programmatic
and presentation of findings at cluster meeting
who will the assessment □ Operational
inform and how you will □ Presentation of findings (e.g. at HCT meeting;
disseminate to inform the □ [Other, Specify] Cluster meeting)
audience □ Website Dissemination (Relief Web, etc.)
□ [Other, Specify]

General Objective [Describe here what the HBDA aims to inform]

Specific Objective(s) [List here what the HBDA aims to identify to facilitate the general objective]

Research Questions [List here the research questions that will need to be answered to meet the objectives]

Geographic Coverage [Describe here the geographic area that the HBDA aims to provide findings about]

Secondary data sources [List here the key secondary data sources that will be included in the research]

Population(s) □ IDPs in camp □ IDPs in informal sites

Select all that apply □ IDPs in host communities □ IDPs [Other, Specify]

□ Refugees in camp □ Refugees in informal sites

□ Refugees in host communities □ Refugees [Other, Specify]

□ Host communities □ [Other, Specify]

Group #: _ _ _ [Other Specify] #: _ _
Stratification Geographical #:_ _ _
Population size per Population size per
Select type(s) and enter □ Population size per strata is □ □
strata is known? strata is known?
number of strata known? □ Yes □ No
□ Yes □ No □ Yes □ No
Data collection tool(s) □ Structured (Quantitative)

Sampling method Data Collection Method

□ Purposive
Structured data □ Household interview (Target #):_ _ _ _ _
collection tool # 1 □ Probability / Simple random
□ Individual interview (Target #):_ _ _ _ _
Select sampling and data □ Probability / Stratified simple random
□ Direct observations (Target #):_ _ _ _ _
collection method and □ Census
specify target # interviews □ [Other, Specify] (Target #):_ _ _ _ _
□ [Other, Specify]

Target level of precision _ _% level of confidence _ _+/- % margin of error

Expected ouput type(s) □ Situation overview #: _ _ □ Report #: _ _ □ Profile #: _ _
Presentation (Preliminary Presentation (Final)
□ □ □ Factsheet #: _ _
findings) #: _ _ #: _ _
□ Interactive dashboard #:_ □ Webmap #: _ _ □ Map #: _ _

□ [Other, Specify] #: _ _

43 43
 Access □ Public (available on humanitarian platforms)
Restricted (bilateral dissemination only upon agreed dissemination list, no publication on
□
other platforms)

Rationale [One paragraph]

• Context
• Information gap
Methodology Overview [One paragraph]
• Summarise each step in the methodology
• Outline any key definitions relevant to the research
Population of Interest [One paragraph]
Explain the rationale for the:
• Geographical area assessed
• Population assessed
• Unit of measurement
Secondary Data Review (outline key bibliography/sources you will use and for what). [One to two paragraphs]
Consider secondary data that will:
• Aid understanding of context
• Identify information gaps – mark the indicators in your data analysis plan that will be addressed through secondary data!
• Key definitions
• Inform assessment methodology including sampling design
• Inform questionnaire design
• Verify/triangulate primary data and findings
Primary Data Collection [Detailed information]
• Method – summarise who is collecting the data, how, when, where, from who and the rationale for this.
• Sampling – summarise your sample for all strata in a table and explain how you will implement the sampling i.e. how you will
select your respondents / participants.
• Tools – explain what tools you will use and how they will be designed and built.
• Triangulation / briefing and debriefing of enumerators – explain how incoming data will be monitored / triangulated and
enumerators briefed/ debriefed.
Data Processing & Analysis [One paragraph]
• Data entry and cleaning process
• Data analysis process – how will you produce the analysis in your Data Analysis Plan

44 44
Annex 6: HBDA Hardware and Software Requirements
A. Hardware
A.1 Laptop computers
A.1.1 Standard laptop package for Assessment Officers
Laptop specifications:
• Processor: i7 quad core (7xxx) or (8xxx), i5 quad core (7xxx) or (8xxx)
• RAM: 8GB - 16GB
• Storage: 256GB - 512GB (must be a solid-state drive)
• Graphics card: Intel HD Graphics 620 or similar
• Display: 14’’-15” (in specific cases and if justification is provided, 13’’ models can also be considered)
• Minimum 1-year manufacturer’s warranty
• Laptop battery life: > 4h
• Laptop battery life (countries with serious electricity supply problems): > 8h
• Processor cores: 4
Accessories:
• Laptop bag
• Wired mouse
• USB-Hub 4 ports (if less than 3 ports on the device)
• Ethernet adapter (if not present on the device)
• Extra charger (countries with serious electricity supply problems)
Potential models:
• Lenovo ThinkPad T4xx (480, 480s, 480p, 470,…)
• Dell Latitude 7xxx (7490)
A.1.2 Standard laptop package for GIS Officers
Laptop specifications:
• Processor: i7 quad core (7xxx) or (8xxx)
• RAM: 16GB - 32GB
• Storage: 512GB to 1TB (must be a solid-state drive)
• Graphics card: Nvidia GeForce GTX 1050 or similar
• Display: 15’’ (in specific cases and if justification is provided, 13’’ models can also be considered)
• Minimum 1-year manufacturer’s warranty
• Laptop battery life: > 4h
• Laptop battery life (countries with serious electricity supply problems): > 8h
• Processor cores: 4-6
Accessories:
• Laptop bag
• Wired mouse
• USB-Hub 4 ports (if less than 3 ports on the device)
• Ethernet adapter (if not present on the device)
• Extra charger (countries with serious electricity supply problems)

45 45
Potential models:
• Dell XPS 15 (indicative public price: 1900-2000 CHF)
• Lenovo ThinkPad Pxxx (indicative public price: 2100 CHF);
• Dell Precision Mobile Workstation (indicative public price: 2200-2500 CHF);

A.2 Mobile Devices

There are many options available for suitable mobile devices. Originally, OpenDataKit, and later KoboToolbox,
applications were designed for use only on Android devices. However, it is now possible to collect data on iOS devices
through web forms. The KoboCollect app can run on any device with Android 2.1 (which was released on January
2010) or above. Note: For Survey 123, the Android version needs to be more recent, 5.0 Lollipop or later (for ARMv7
32 bit) or 6.0 Marshmallow or later (for ARMv8 64 bit).
When selecting a device you need to consider a number of factors:
• Cost
• Durability
• Battery life
• GPS accuracy
• Screen quality
• Connectivity options
• Phone vs. Tablet
In practice, it is good to try to achieve a balance between acceptable pricing (this generally rules out Apple devices)
and choosing devices that can be re-used in the field for multiple deployments. Phones that are unlocked and that use
multiple frequencies (quad band) are ideal. Mid-range devices make the most sense, as you can avoid the pitfalls of
the lowest tier devices as well as the unnecessary and costly features of the higher end devices.
More information from the KoboToolbox team can be found here and regarding Survey 123, here.

A.3 Supplemental GPS/GNSS receiver (optional)

In cases where the GPS accuracy of mobile devices is not good enough, there are options to provide real-time submeter
GPS corrections in the field. GPS/GNSS receivers are able to work with any mobile device via Bluetooth connection
and can work with any third-party or in-house data-collection app.
The Arrow 100 from EOS Positioning Systems represents one of the best options on the market.

B. Software
B.1 Essential Software
B.1.1 Data Collection
• Option1: KoboToolbox – recommend as the default option, ideally using the KoboCollect app installed on
Android devices.
• Option2: OpenDataKit (ODK) – recommended for situations when there will be little to no connectivity in
country, the fully offline workflow combining ODK Collect and ODK Briefcase has proved reliable.
• Option3: Survey 123 for ArcGIS – recommended for situations where additional functionality provided by the
ESRI suite of ArcGIS field applications justify the cost for ArcGIS Online subscriptions. The combination of

46 46
 Workforce, Navigator and Survey 123 appears promising. Further the integration with Operations Dashboard
could add significant value as well.
B.1.2 Data Visualization and Analysis and Reporting
• Microsoft PowerBI – proven to be a useful tool for visualizing and analyzing HBDA datasets
• Microsoft Office – Excel and Word are industry standard
B.1.3 Antivirus
• Bitdefender or similar
B.2 Optional Software
• ArcGIS Desktop or Pro – for more advanced geospatial applications / analysis
• Adobe Creative Cloud – for enhanced graphic document production / desktop publishing

47 47
Annex 7: UNDP-UNOSAT Standard Operating Procedures

UNDP’s engagement in crisis preparedness and response can be supported by satellite-based analysis and
geoinformation, crucial for timely decision-making on the allocation of resources and prioritization of activities.

In October 2017, UNDP and the Operational Satellite Application Programme (UNOSAT) of the United Nations
Institute for Training and Research (UNITAR) signed a Standard Operating Procedure (SOP) to streamline the use
of geospatial technologies for UNDP Country Offices and Regional Hubs for emergency and crisis response,
early warning and preparedness, risk assessments and recovery planning at country and regional level.

HOW TO ACCESS UNOSAT SERVICES

UNOSAT products and services include satellite imagery, maps, geospatial analysis, deployment of experts and
training, and can be accessed through two modalities:

RAPID MAPPING
During humanitarian emergencies – both natural disasters and conflicts – UNDP can activate at no cost UNOSAT
Rapid Mapping Service, and can access directly from its website products/analyses relevant by area and by
crisis, unless restricted by copyright or other limitations.
See this short introductory video and UNOSAT Geospatial Catalogue for examples of satellite-imagery-
derived products that the Country Office can request to support immediate crisis response.

PAYING SERVICES
UNDP will pay for the following UNOSAT services:
• In-country support, including training and deployement of GIS experts and/or geospatial coordinators at
short notice.
• Support on risk assessments and recovery planning at country level.
• Other kinds of technical support involving assessments, monitoring over time of recovery and development
projects, and the development of Post Disaster Needs Assessments (PDNA) and Recovery and
Peacebuilding Assessments (RPBA).

EXAMPLES OF UNOSAT – UNDP COOPERATION

The satellite-based damage assessment in Gaza - OPT has determined the location of the most affected
communities, provided a preliminary infrastructure damage assessment, and supported planning of a detailed ground-
based damage assessment.

UNDP activated UNOSAT Rapid Mapping Service for achieving situational awareness and assessing the impact of
floods in Vietnam, Uganda, and Guatemala in 2017.

48 48
HOW TO REQUEST UNOSAT SERVICES – NO COST AND PAYING OPTIONS
1. Contact the Regional Bureau desk and the Crisis Bureau Country Support Management Team (CSMT) for
immediate support on mapping and other key information needs. Based on the initial discussion with the Country
Office, CSMT submits an immediate request to UNOSAT with copy to the relevant Country Office and Regional
Bureau counterpart, and Crisis Bureau in Geneva. UNOSAT evaluates the feasibility of the request and reverts to
UNDP within 24 hours with a delivery time for the product/service.
2. In case the Country Office does not have the capacity to make an immediate request, CSMT in consultation with
the Regional Bureau desk, sends an initial request to UNOSAT and copies the relevant Country Office counterpart
and Crisis Bureau in Geneva.
3. No specific format is required for submitting requests, but the following information should be included:

a. Type of information needed – for example impact of tropical storm, extent of flood, assessment of
damages to infrastructure, estimation of population exposed to a disaster, etcetera.

b. Area of interest/location – ideally sent as KML file or latitude/longitude coordinates, however location
names are also acceptable.
4. UNOSAT makes available satellite-imagery-derived products for digital download in different formats: PDF,
GIS-ready data, Maps - Live Maps, Reports & Statistics.
In the case in which paying services are requested UNOSAT provides a cost estimation. Procurement of paying
services will be arranged through a UN to UN agreement. Funding options include payment from the requesting
entity, Crisis Bureau seed funding and joint project development/resource mobilization with UNOSAT.

WHY UNOSAT SERVICES ARE STRATEGIC TO UNDP’S COUNTRY OFFICES

Satellite based analysis and geospatial information are key to complement rapid assessments carried out by the
Humanitarian Country Team on the ground with concrete figures of damages that are often not available in the first
days, particularly when access to some areas is difficult or not possible.

FOCAL POINTS

Crisis Bureau – Country Support Management Team, New York

Africa :
• Aboubacar Koulibaly: [email protected];

Asia and Pacific, Europe and CIS:

• Rebecca Reynolds: [email protected];

Arab States:
• Aurelie Boukobza: [email protected]

Crisis Bureau - Geneva

Roberto Paganini: [email protected];

• UNOSAT – Geneva
Olivier Van Damme: [email protected]; ; [email protected].

49 49