Reproducibility in Social Science: Two lectures for Endangered Data Week [UPDATE: Lecture #2 is postponed]

Endangered Data Week is a new, collaborative effort, coordinated across campuses, nonprofits, libraries, citizen science initiatives, and cultural heritage institutions, to shed light on public datasets that are in danger of being deleted, repressed, mishandled, or lost. The week’s events can promote care for endangered collections by: publicizing the availability of datasets; increasing critical engagement with them, including through visualization and analysis; and by encouraging political activism for open data policies and the fostering of data skills through workshops on curation, documentation and discovery, improved access, and preservation. Learn more about Endangered Data Week. 

The following events at Yale, hosted by The Policy Lab at the Institution for Social and Policy Studies, may be of interest to the YLS community. As always, feel free to contact Empirical Services Librarian, Michelle Hudson, for assistance with your empirical legal research projects, or to discuss issues related to reproducibility, data availability, and preservation.

Why Reproducibility in (Social) Science Matters (and How to Get it Right)

Thursday, March 1, 2018
10:30-12:00 
ISPS Policy Lab, 77 Prospect Street

This talk will give an overview of the relevant history and philosophy of science with respect to reproducibility, mostly using examples from psychology, and explaining why reproducibility is so important. Then there will be a discussion of how to properly conduct and interpret replication studies, stressing that it’s not straightforward but requires painstaking care.

Speaker: Brian Earp is a PhD candidate in the Departments of Philosophy and Psychology, Yale University. He is Associate Director of the Yale-Hastings Program in Ethics & Health Policy, and Research Fellow at the Uehiro Centre for Practical Ethics, University of Oxford. Brian’s work is cross-disciplinary, following training in philosophy, cognitive science, psychology, history and sociology of science and medicine, and ethics. Brian is the recipient of both the Robert G. Crowder Prize in Psychology and the Ledyard Cogswell Award for Citizenship from Yale University, where he was elected President of the Yale Philosophy Society as an undergraduate as well as Editor-in-Chief of the Yale Philosophy Review. He then conducted graduate research in psychological methods as a Henry Fellow of New College at the University of Oxford. While at Oxford, he completed additional coursework in the philosophy of science and philosophy of mind, which he went on to publish in peer-reviewed journals. He also conducted graduate research in the history, philosophy, and sociology of science, technology, and medicine as a Cambridge Trust Scholar and Rausing Award recipient at Trinity College at the University of Cambridge. He is currently a Gordon Fellow, Benjamin Franklin Graduate Fellow, and McDougal Writing Fellow at Yale University conducting doctoral research in philosophy and psychology, having been jointly admitted to both departments. His essays have been translated into Polish, German, Italian, Spanish, French, Portuguese, and Hebrew.

Yale co-sponsors: ISPS, Yale Day of Data, Center for Science and Social Science Information, Graduate Writing Lab

The lecture by Florio Arguillas on Friday, March 2, has been postponed due to inclement weather. I’ll post a new news item when the talk is rescheduled. 

Making Research Transparent and Reproducible

Location: ISPS Policy Lab, 77 Prospect Street

Replication of results is a core requirement of the scientific method. Satisfying this requirement becomes increasingly complex when data from disparate sources is integrated and reused. While code used for analysis of data must be verified, it is also imperative that code and processes used to clean, integrate and harmonize data be documented and verified. This can be time-intensive and intimidating, especially for individual researchers seeking to openly share their work. Workshop participants will be introduced to CISER’s Reproduction of Results Service workflow which incorporates the Teaching Integrity in Empirical Research (TIER) Protocol.

Speaker: Florio Arguillas is a Research Associate at the Cornell Institute for Social and Economic Research (CISER). Florio conducts workshops and works with individual researchers at Cornell on the management and processing of qualitative and quantitative data, stewardship of private or confidential data, and the use of statistical software packages.  As a consultant with the Research Data Management Service Group, he helps researchers develop data management plans and access the services necessary to implement them. Florio established CISER’s Data Curation and Reproduction of Results Service, or R2, which allows researchers to submit their data and code to CISER prior to manuscript submission for appraisal, curation, and replication by CISER data curation and computing experts. He also co-founded Curation for Reproducibility (CURE), a consortium of Cornell, Yale, and the University of North Carolina, that promotes workflows and documentation methods that enhance the reproducibility of statistical research. As a TIER Fellow, Florio incorporates the TIER Protocol in the trainings on Research Transparency and Reproducibility that he conducts for several cohorts of Cornell graduate students every year. Florio earned an A.B. in Economics at the Ateneo de Davao University, an M.A. in Demography at the University of the Philippines-Diliman, and  both an M.S. and Ph.D. in Development Sociology at Cornell University.

Yale co-sponsors: ISPS, Stat Lab, Center for Science and Social Science Information, Yale Center for Research Computing