2016 : WHAT DO YOU CONSIDER THE MOST INTERESTING RECENT [SCIENTIFIC] NEWS? WHAT MAKES IT IMPORTANT?

Sara Miller McCune Director, Center For Advanced Study in Behavioral Sciences, professor, Stanford University; Jere L. Bacharach Professor Emerita of International Studies, University of Washington
Big Data And Better Government

Big data gives business, government, and social scientists access to information never available before. With the right tools of analysis—which are improving exponentially as I write—big data will transform the way we understand the world and the means we use to fix problems. The US and other governments are building the capacity to use big data as a basis for determining best practices; university-based research programs are generating appropriate analytic tools; and various non-profits around the world are linking technology, data, and citizens to enhance the implementation of government programs and services.

Science can now effectively be brought to bear on public policymaking. Yet, important distinctions exist among the key players. One set of actors wants to ensure that public policies are evidence-based, and a second set aims to enable citizens to complain about poor services and to get the services they need. Some are fundamentally concerned with the science and others with voice.

Evidence-based policy has become a mantra in some circles, and increasingly the focus is on assessment of policies once enacted as well as on the ex ante crafting of good policy. Randomized experiments have gained popularity worldwide by bringing scientific rigor into the appraisal of interventions meant to improve well-being. But they are not the only tools in the toolbox. Observational analyses using big data are just as important, particularly where randomization of people and communities is undesirable, infeasible, unethical, inadequate, or all of the above. Political considerations often trump randomization when it comes to the location of hospital facilities, military bases, and schools. Even in very politicized circumstances, new techniques of causal inference from observational data make it possible to learn about the conditions under which different policies are likely to succeed. Indeed, the progress in recent years on generating scientific inferences from observational data has been breathtaking. 

Simultaneously, another group of actors are stepping up to the plate to adapt and improve current technologies, data platforms, and analytic advances in the service of citizen voice. Providing individuals with mobile phones to take pictures, send texts and emails, and otherwise document what they see offers citizens a means for reporting on where things are broken and for demanding that they be fixed. It is also a new and important form of quality control over elections, services, and bureaucrats. Reporting leaking gas mains or water hydrants, photographing pot holes and abandoned homes, and naming corrupt officials can lead to significantly improved government responsiveness—and in some places already has, generally as a result of the work of nonprofits, such as Code for America in the US and eGovernments Foundation in India, or of university-based research teams collecting evidence on how government actually functions. One recent success involves discovering and correcting the gap in the distribution and use of food stamps in California.

The amount and kind of data collected from all of us does pose dangers to privacy and misuse. Science and engineering are being mobilized to assure the proper protections are in place, but governments must also convince publics that they are trustworthy in how they use the data that they access. At stake is the promise of better government that draws on scientific analysis of policy and scientific and technological amplifiers of voice.