Remote Data Collection in Fragile Contexts
In fragile and conflict-affected settings, traditional methods of Monitoring and Evaluation (M&E) often break down. Enumerators can’t reach communities, infrastructure is damaged, and security risks make on-the-ground data collection impractical or even dangerous. Yet, the need for reliable, timely information in these settings is more critical than ever.
Thanks to technology, M&E professionals now have an evolving toolbox for collecting data remotely. Whether it’s through mobile phones, satellite imagery, or artificial intelligence, remote data collection is enabling programs to continue monitoring and learning—even in the most constrained environments.
Why Remote Data Collection Matters
Fragile contexts are marked by political instability, natural disasters, conflict, or displacement. Data helps humanitarian and development actors understand needs, monitor progress, and adapt interventions quickly. But how do you gather that data without putting people at risk?
Remote methods offer:
- Safety for enumerators and respondents
- Continuity of monitoring even during crises
- Broader reach across hard-to-access areas
- Real-time feedback for adaptive programming
1. Mobile-Based Surveys
Tool Examples:
- KoboToolbox
- SurveyCTO
- RapidPro
Approach: Surveys are administered via SMS, phone calls, or mobile apps.
Pros:
- Fast to deploy
- Supports two-way communication
- Works offline in many cases
Challenges:
- Phone ownership and literacy barriers
- Potential sampling bias (excludes those without phones)
Field Insight: In South Sudan, NGOs used interactive voice response (IVR) to collect feedback from displaced communities, minimizing the need for enumerator presence.
2. Satellite and Remote Sensing
Tool Examples:
- Google Earth Engine
- Sentinel Hub
Approach: Uses satellite imagery to assess infrastructure, environmental changes, or population movements.
Pros:
- No need for physical presence
- Covers vast areas quickly
- Objective and repeatable
Challenges:
- Limited to visible or environmental indicators
- Requires GIS and image analysis skills
Field Insight: In Afghanistan, remote sensing was used to monitor the construction of schools and roads for USAID projects, verifying progress without entering high-risk zones.
3. Chatbots and AI Assistants
Tool Examples:
- RapidPro
- Viamo
- Facebook Messenger Bots
Approach: Uses automated chat interfaces to ask questions, collect feedback, or deliver messages.
Pros:
- 24/7 accessibility
- Multilingual support
- Scalable to thousands of users
Challenges:
- Requires internet or mobile data
- Limited depth of conversation
Field Insight: In Haiti, a chatbot was deployed via WhatsApp to collect rapid assessments after an earthquake, reaching over 20,000 users within a week.
4. Crowdsourced Data
Tool Examples:
- Ushahidi
- OpenStreetMap
- Humanitarian Data Exchange (HDX)
Approach: Aggregates data from community members, volunteers, or online contributors.
Pros:
- Empowers local voices
- Real-time situational awareness
- Useful in emergencies and conflict zones
Challenges:
- Data reliability and verification
- Risk of misinformation or duplicate reporting
Field Insight: During the Ebola crisis in West Africa, Ushahidi was used to crowdsource reports of symptoms, helping responders identify outbreak hotspots faster.
Best Practices for Remote Data Collection
| Principle | Why It Matters |
|---|---|
| Ethics First | Always prioritize informed consent, especially when remote methods limit oversight. |
| Data Triangulation | Combine multiple sources (e.g., phone, satellite, admin data) to validate findings. |
| Tech Inclusivity | Ensure tools are accessible to marginalized groups (e.g., women, non-literate users). |
| Remote Enumerator Training | Equip staff with guidance on conducting phone or virtual interviews ethically and effectively. |
| Local Partnerships | Work with community-based organizations to extend reach and trust. |
What the Future Holds
AI is beginning to play a larger role in analyzing remote data—whether through natural language processing of open-ended SMS responses or machine learning to detect anomalies in remote sensing. As these tools become more user-friendly, their adoption in fragile contexts is likely to grow.
Final Thoughts
Remote data collection is no longer a workaround—it’s becoming a standard part of the M&E toolbox. When done right, it ensures programs stay informed, accountable, and responsive even when access is limited. With careful planning, the right tech, and local collaboration, evaluators can keep learning alive in the world’s most challenging environments.
Want templates, remote training guides, or sample phone survey scripts? Check out EvalCommunity Academy Premium.
