Data challenges in global education: The limits of what we know

It is universally agreed that in order to solve the world’s most intractable problems, we need reliable metrics to understand their scope and scale and to track progress in finding solutions. The calls for a “data revolution” from Bill Gates and others have been loud and clear, and the need for a response in education is evident.

FHI 360’s Education Policy and Data Center (EPDC) is working hard to ensure that any such education data revolution includes a substantive movement toward addressing data gaps. Most recently, we provided country and subnational estimates of the number of out-of-school children, using household survey data. These numbers filled in the gaps where no existing estimates were available, which helped our programs and partners craft evidence-based responses.

Often, large-scale data estimates are taken at face value. For instance, it is commonly believed that 57 million primary-aged children were out of school in 2011. However, EPDC and more recently, UNESCO through its 2013/14 global monitoring report, have stated that this figure is not precise and that there is likely a significant amount of variability and missing data.

At the Comparative and International Education Society conference in Toronto this week, data experts from FHI 360’s Global Learning team will discuss the common challenges and limitations inherent in large-scale data collection and dissemination in a panel session, The limits of what we know: Missing data in education and development. We will explore some of the techniques used by organizations — including the UNESCO Institute for Statistics, the International Labor Organization and the World Bank — to address data gaps and show how strategies to address missing data may be used and combined to produce an estimate of out-of-school children in sub-Saharan Africa that ranges from 26 to 36 million.

As we approach the 2015 deadline for the Education for All goals, it is critical to recognize that the best the international community may be able to do at this point is to provide ranges of estimates for issues such as out-of-school children. While useful for stimulating interest or simplifying a newspaper headline, single numbers — such as 57 million children out of school or 250 million children in school but unable to read — often convey a false sense of precision and are inadequate for real development policy and planning.

An important component of any data revolution in education will be to limit the data ranges we use and come to a shared understanding of how to define and address global development challenges accordingly.

No Responses

More From This Series