TL;DR
This paper explores the feasibility of using mobile phones for population health surveys in low-resource areas, specifically the Philippines. They developed and tested an offline-capable mobile form software that tracks survey completion times, demonstrating its viability and improving surveyor preference for mobile over paper surveys.
What Problem Does This Paper Address?
Conducting large-scale population health surveys is crucial for effectively allocating resources in low-resource communities. However, traditional pen-and-paper methods are labor-intensive for data tabulation and analysis. While digital surveys are common in affluent regions, the cost of dedicated devices and the need for internet connectivity pose significant barriers in low-resource settings. Existing mobile survey solutions often fail to meet the specific needs of these environments, such as offline functionality, ease of use for non-tech-savvy users, and the ability to track survey efficiency metrics.
Methods
The researchers employed a multi-stage approach:
- Pilot Interviews: They conducted interviews with potential volunteer surveyors to understand their comfort level with technology, phone usage patterns, and preferences for survey methods (paper vs. mobile). This phase also helped identify key requirements for a mobile survey tool.
- Software Development: Based on pilot interview feedback, they developed custom software to generate mobile forms. Key features included:
- Offline Capability: Essential for areas with unreliable internet.
- Ease of Use: Designed with a single question per screen and minimal visual complexity to reduce training time and cognitive load.
- Time Tracking: The software automatically records time spent on each question, typing time, and overall survey completion time.
- Field Testing: The developed mobile form software was deployed in three diverse locations in Metro Manila, Philippines. 33 volunteer surveyors collected health survey data from 266 respondents.
- Post-Field Testing Interviews: After the field deployment, surveyors were interviewed again to assess their user experience, satisfaction, and overall preference for mobile versus paper surveys.
Datasets
Not discussed in the paper. The study utilized data collected from 266 patient surveys during the field testing phase.
Key Insights and Ideas
- Preference Shift: Initial reluctance towards mobile forms due to perceived difficulty and internet dependency can be overcome with hands-on experience. The surveyors' preference for mobile forms significantly increased after using the developed software.
- Offline is Non-Negotiable: The field testing revealed that internet connectivity was unavailable 25% of the time, strongly validating the necessity of offline functionality for mobile surveys in these regions.
- Simplicity Drives Adoption: A user interface that presents one question at a time, minimizes visual clutter, and requires minimal training is critical for engaging surveyors who may not be highly tech-literate.
- Data-Driven Efficiency: The ability to track time spent on individual questions provides valuable insights for optimizing survey design, identifying complex questions, and potentially framing them as multiple-choice for faster completion.
Key Results
- Pilot Preference: In pilot interviews, 60% of surveyors preferred mobile forms over paper.
- Training Efficiency: 96% of surveyors completed training in 10 minutes or less.
- Survey Completion: 266 population health surveys were successfully collected by 33 surveyors.
- Post-Field Preference: After field testing, the preference for mobile forms increased to 76%.
- Ease of Use: 70% of surveyors found the mobile form "very easy" to complete.
- Typing Speed: Average typing speed was 11.5 words per minute, with significant variation.
- Offline Usage: Internet was unavailable 25% of the time during surveys, highlighting the importance of offline capability.
- Question Type Efficiency: Multiple-choice questions generally took less time to answer than text-based questions.
Limitations and Future Work
- Limited Usage: Surveyors only used the mobile form an average of 8 times during the field testing. The authors hypothesize that increased usage would further solidify preference for mobile forms.
- Generalizability: While tested in the Philippines, the findings' applicability to other low-resource settings with different cultural contexts and technological penetration may vary.
- Future Work: The authors suggest that with more usage, preference for mobile forms would continue to increase. Further research could explore the long-term impact of mobile surveys on data quality and cost-effectiveness compared to traditional methods. Optimizing survey design based on time-tracking data is also a promising avenue.
Summary updated