Mit Cms/w

How will the role of data science in democracy be transformed as software expands the public’s ability to conduct our own experiments at scale? In the 1940s-70s, debates over authoritarian uses of statistics led to new paradigms in social psychology, management theory, and policy evaluation. Today, large-scale social experiments and predictive modeling are reviving these debates. Technology platforms now conduct hundreds of undisclosed experiments per day on pricing and advertising, and the algorithms that shape our social lives remain opaque to to the public. Democratic methods for data science may offer an alternative to this corporate libertarian paternalism. In this talk, hear about the history and future of democratic social experimentation, from Kurt Lewin and Karl Popper to Donald Campbell. You’ll also hear about CivilServant, software that supports communities to conduct their own experiments on algorithms and social behavior online. J. Nathan Matias is a Ph.D. candidate at the MIT Media Lab Center