Session: University-Agency Data Partnerships: Candid Reflections and Lessons Learned (Society for Social Work and Research 21st Annual Conference - Ensure Healthy Development for all Youth)

310 University-Agency Data Partnerships: Candid Reflections and Lessons Learned

Schedule:
Sunday, January 15, 2017: 11:30 AM-1:00 PM
Bacchus (New Orleans Marriott)
Cluster: Research Design and Measurement
Speakers/Presenters:
Emily Putnam-Hornstein, PhD, University of Southern California, Daniel Webster, PhD, University of California, Berkeley, Regan Foust, PhD, University of Southern California, Shraddha Deshmukh, MS, University of Southern California, Jonathan Hoonhout, BA, University of Southern California and Wendy Wiegmann, MSW, University of California, Berkeley
Conversations about the value of cross-sector record linkage as a means to reducing redundancies and more effectively coordinating services are decades old. Yet significant effort is required to curate data originally generated for non-research purposes. Even though federal and state laws permit the use of most administrative records for research and evaluation efforts, agencies are often hesitant to share data with university researchers for analysis. Separately, from a scientific standpoint, technical documentation that would allow researchers to accurately analyze agency-generated data is often missing—and tools for matching client information across data sources are either cost-prohibitive or poorly optimized for the task. In this roundtable, panelists from USC’s Children’s Data Network and UC Berkeley’s Child Welfare Indicators Project will share lessons learned and candid reflections about university-agency data and research partnerships. Topics will include: (1) supporting researchers in the analysis of administrative records through meta-data documentation, code repositories, mentorship, and peer-review; (2) engaging non-academic stakeholders in data by using animations, infographics, and other translation tools; (3) record linkage and the customization of open-source, machine-learning algorithms and technologies that identify, match and probabilistically link records; (4) publishing aggregated statistics on open data portals and providing technical assistance; (5) best practices for data security and record confidentiality; (6) federal statutes governing research activities with administrative data; and (7) handling media inquiries to balance intellectual freedom with political sensitivities. The roundtable will be structured as an active Q&A with panelists and audience members alike contributing to the conversation.
See more of: Roundtables