The Science of Data Science - Developing a Data Framework and Methods to Bring the All Data Revolution to Communities

• Activity Number: 157 - Constructing Profiles of Local Communities
• Type: Topic Contributed
• Date/Time: Monday, July 30, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Stephanie Shipp* and Sallie Keller
• Companies: Biocomplexity Institute of Virginia Tech and Social & Decisional Analytics Lab, Virginia Tech
• Keywords: data discovery; data profiling; data linkage; data integration; fitness-for-use; community-based research

Abstract

Existing data flows at local levels are ubiquitous in our everyday life. These data include administrative records, geospatial data, social media, and surveys. Communities would like to use these data to address their challenges in delivering services and taking care of their vulnerable populations. At the same time, the federal statistical community is interested in making use of local data flows to enhance and potentially reduce collection of survey data to provide timelier and more geographically detailed data. Yet, these local data sources are not designed for statistical analysis and methods are lacking to use these data in a statistically rigorous way. Through our research with local, state, and federal agencies, we are developing a data framework to create methods to discover, screen, profile, explore, assess their fitness-for-use, and integrate multiple data sources for use to address specific research questions and support evidence-based decision making. Our data framework encapsulates a systematic approach based on statistical and social science principles. Drawing synergies out of a collection of these case studies is critical for maturing the Science of Data Science.