Traditional usability methods are of limited use when evaluating systems designed for distant, diverse populations. In this paper, we describe a study conducted in two Ghanaian villages that evaluated an audio computer designed for people living in oral cultures. Informed by ICTD and orality-grounded HCID, we modified existing usability testing practices and we reflect on the utility of these adaptations. We found that conducting a culturally appropriate study often meant forgoing more traditional approaches in favor of flexible, opportunistic methods. We acknowledge the challenges of adapting traditional usability methods for oral, rural users. However, we found that by implementing strategic modifications led by local staff, our study produced valuable, actionable results.

Technology projects are finding ways to provide information to people living in rural poverty. However, using information to affect health or farming practices requires overcoming unique challenges including illiteracy and lack of electricity. We examine the effects of a low-cost audio computer ("Talking Book")--a battery-powered, durable, handheld device that enables users to create and listen to recordings and copy recordings between devices-for improving learning opportunities and knowledge sharing in such environments. In northern Ghana, we studied the impact of giving rural, illiterate people on-demand access to guidance created by local experts. Our evaluation suggests that Talking Books can make a significant impact on learning and behavior change in villages with low literacy rates and no electricity.