I recently taught a 8 week course in R at the Centro de Investigação em Saude de Manhiça, a health research facility in Southern Mozambique. The class was a huge success: enthusiastic students, a fun and collaborative class atmosphere, and lots of data.

The students learned a ton, but so did I (the teacher). Specific to the subject of this post, my lessons learned were that:

• Licensing fees and processes pose an even bigger burden on researchers in the African context than in the U.S., Canada, or Europe.
• Most students had almost no "formal" access software (through an institution) or materials for learning about software.
• Most students were familiar with at least one commercial statistical software (Stata, SPSS, Excel, SAS), and many had a copy (often pirated) of one.
• Most students' familiarity with these statistical softwares was relatively low, mainly due to a lack of guidance/teaching.
• Many students were surprised by (a) the fact that R is free, and (b) the fact that so much can be done in R (maps, web apps, etc.).
• Nearly all students said they'll continue working in R after the course ends.

What's the take-away?

Open source software has the potential to have a major positive impact on productivity and innovation in the sphere of African health research. But there are two pre-requisites to this transformation taking place: more exposure to "free" tools at the institional level, and more education resources for languages like R and python at the individual level.