Social Data Collaboratory

The Social Data Collaboratory (SDC) uses social media and data science to inform decisions.

The internet and social media have redefined how we search for, are exposed to, produce and share knowledge and ideas. These evolving forms of communication, combined with traditional media, saturate our society with messages that are powerful and complex. This dynamic has profound implications and equally profound opportunities for studying attitudes and behaviors related to health, healthcare, politics, education, and a host of other subjects. Our team works with numerous types of digital media and social data.

We bring together social science researchers, data and computer scientists, and new media specialists to collaborate across and beyond academia.

SDC uses rigorous, innovative and multidisciplinary methods to study the intersection of digital communications and human opinions and behavior. We examine how information is searched for, engaged with and shared on traditional and new forms of media. We measure the amount and diffusion of online posts and analyze the content and sources of messages to understand their effects on public opinion and behaviors. Our methodologies draw on established frameworks from applied statistics and computational linguistics and employ cutting-edge machine learning techniques. We create our own surveys to supplement our understanding of how individuals use media and how this experience shapes their attitudes, beliefs and behaviors regarding various topics of interest.

We are committed to setting transparent methodological standards, and sharing what we learn.

SDC has developed disclosure standards to document every step of the social media data acquisition and management process: keyword precision and recall, data parsing, feature extraction and selection, human coding, classifier development and performance, and resulting data. SDC strives for transparency and replicability of social data research. We share knowledge and research findings with policymakers, the research community and the public so that we all can make informed decisions that protect and promote our health.