Impact Analysis & Mobile Data Solutions

Click here to read the first post in my series on expanding my education through MOOCs.

As an international development professional, I often joke that I have no marketable skills. An exaggeration, of course, but it's hard not to feel that way when surrounded by software developers and statisticians. Therefore, part of my professional development journey has been to improve my skills related to ICT4D -- information and communication technologies for development -- and impact analysis.

I recently completed two courses that complemented this goal nicely: +Acumen's Making Sense of Social Impact: Building Blocks of Impact Analysis and TechChange's Mobile Data Solutions.

Making Sense of Social Impact

Acumen is a non-profit venture fund that invests in companies, leaders, and ideas that address global poverty. They call this "social impact investing." With this worthy goal, they've developed an approach to ensure that their investments are delivering as promised.

They begin with a problem, such as the limited availability of sustainable lighting for the poor in developing countries. The theory of change is that affordable, portable solar lanterns will eliminate the reliance on kerosene, thereby reducing instances of lung disease, mitigating carbon emissions, and decreasing household spending over time. (Click here to learn more about solar products company d.light.)

To arrive at this proposed solution, d.light had to learn about its target market and unique local challenges. Even after the solar lanterns were distributed, d.light continued to measure whether the input (lantern distribution) contributed to the desired outcome (reduced kerosene use, reduced indoor pollution) and long-term impact (fewer instances of lung disease, reduced carbon emissions).

This is where efficient and effective surveys are of utmost importance.

Mobile Data Solutions

In the old days of global development, enumerators would travel home-to-home with a stack of papers and a pen, interviewing subjects who spoke numerous languages and dialects. Transcription could take months and produce costly errors.

Today, organizations increasingly rely upon mobile devices to quickly and transparently collect data about people, projects, and programs.

For example, in Uganda, young people use their own mobile devices to participate in weekly SMS surveys to share information about disease outbreaks or inform public policy. In other cases, enumerators use their mobile devices to record answers during traditional home-to-home surveys, but instead of waiting several months for data to be transcribed, the data can be immediately exported for analysis.

I experimented with textit.in, a service that allows users to build SMS surveys for free:

In addition to tips for survey design, the course provided a brief introduction to data visualization best practices. Which of these graphs best represents the change in numbers of deaths from communicable diseases relative to other causes of deaths?

Fortunately, I've slowly been working my way through the Edward Tufte books that have been gathering dust in my apartment, so I got this one right. The graph on the left might be more eye-catching, but it obscures the data. It's always preferable to have a simple graph that clearly conveys the significant trends.

So, that's another two classes down! Now, I'm going to try out some of the open-source mapping and data visualization resources shared in the class (now that I'm free from the pressure of getting too creative with my charts).