Open Source Data Collection in the Developing World

Yaw Anokwa, Carl Hartung, et. al. | Invisible Computing | October 15, 2009

ODK is designed to let users own, visualize, and share data without the difficulties of setting up and maintaining servers. The tools are easy to use, deploy, and scale.

In the developed world, data is relatively easy to collect. Be it population demographics, embedded traffic sensors, or even popular Internet services, the ability to easily tap and synthesize raw data enables individuals and organizations to make decisions. Examples of such synthesis include earthquake sensing via Twitter, traffic mapping in Google Maps, and disease-oriented websites like PatientsLikeMe and Google Flu Trends.

In the developing world, the lack of reliable infrastructure, ubiquitous connectivity, and adequate expertise makes data collection difficult. Currently, most organizations collect data on paper forms despite inefficiencies such as the physical collection of completed forms, data transcription errors, and long delays before the data is available. 

This problem is exacerbated by the data’s critical nature. If, for example, you don’t know how far villagers are from a stagnant water source, it’s difficult to know how many mosquito nets to deploy; and, if deployment information isn’t connected to malaria cases at local clinics, it’s impossible to know whether the nets have made a difference. The exponential growth of cell phone usage and infrastructure in developing regions has aroused great excitement for using mobile devices to address current gaps in data gathering.

In addition to the variety of data—text, photos, location, audio, video, barcode scans—that can be gathered, mobile devices have proven to be dramatically faster at both collecting the data and making it available to decision makers. Moreover, deploying mobile devices can be less expensive and less error prone than using pen and paper.