Any kind of data may be less objectively interpretable than it may seem. Scientific processes and “facts” have always been created by humans who exist in power structures and have biases. Creating quantitative data that cover an entire population is a good step toward capturing relevant evidence about all groups, but we should still be careful to understand the biases we use to interpret data points and categories they’re put in. Dr. Mareike Schomerus is Vice President at the Busara Center in Nairobi. In this video, she shares her experiences with practical ways to design research and use data in ways that are both good science as well as grounded in local reality.