Using big data for impact evaluations

Using big data for impact evaluations

This Methods Brief by the Centre of Excellence for Development Impact and Learning (CEDIL) takes a step-by-step, practical approach to guide researchers designing impact evaluations based on big data. 

The amount of big data available has exploded with recent innovations in satellites, sensors, mobile devices, call detail records, social media applications, and digital business records. Big data offers great potential for examining whether programmes and policies work, particularly in contexts where traditional methods of data collection are challenging. During pandemics, conflicts, and humanitarian emergency situations, data collection can be challenging or even impossible.

Big data offers great potential to fulfil data needs and to help establish which policies or interventions work, even in contexts where traditional methods of data collection are challenging. Integrating big data with traditional household surveys and administrative data can improve data availability, quality, granularity, accuracy, and frequency, and can help measure development outcomes over time and space in a number of new ways,