Policy for Big Data: An Investigation Using Land Records

In an increasingly digitised world, Information and Communication Technologies (ICTs) are playing a key role in human development. A consequence of this increased digitisation is the exponential increase in the being created. Data from multiple sources can be “mashed up” and processed using advanced analytics to provide deep and detailed insights on the world around us. However, there are many challenges in the use of “Big Data for Policy”, a key one being around the “Policy for Big Data”. An often understated aspect is the fact that much of this “big data” (especially in the emerging economies) is owned by the private sector that has the necessary wherewithal to collect and analyse it. Nevertheless, during the process of governing, the public sector creates “administrative data”, which is a unique resource. However, even creating administrative data in a form that can be linked together to create “big data” is fraught with challenges. The aim of this research is identify these challenges so that policy makers can attempt to mitigate them. Using the domain of land records, I have created a novel dataset on the proliferation of a land computerisation project in the states of India, seeking to identify the key factors behind the uneven uptake of this project in the states. I hypothesise that the diffusion of policies that seek to create digital records depend on three main factors-—the relative level of socio-economic development, the amount of support a policy would have amongst the populace and the complexity of policy adoption. The research finds significant evidence for all the three hypotheses.