Dec 11, 2014 – Dec 12, 2014 | Berkeley, CA
The movement towards more transparency, reproducibility, and openness has gained a lot momentum in the social sciences. Yet, the norms and institutions that govern academic research do not reflect this culture shift. Significant problems remain, including professional incentives that reward striking and statistically significant research findings at the expense of scientific integrity.
Increasing the reliability and accuracy of scientific evidence requires well-defined standards of methodological rigor. At the same time, new tools and strategies to increase transparency must be integrated into existing research workflows to facilitate adoption. As the social sciences reinvent their practices around data, it is absolutely the right moment to build new channels of collaboration, cross-learning, and dissemination for innovative, open research practices.
The two-day conference brought together academic leaders, scholarly publishers, and policy-makers to discuss recent innovations in journal practices, academic training, data sharing, and evidence-based policy in light of the push for increased transparency.
Neil Malhotra (Stanford University): “Publication Bias in the Social Sciences: Unlocking the File Drawer”
Uri Simonsohn (University of Pennsylvania): “False-positive Economics”
Maya Petersen (UC Berkeley): “Data-adaptive Pre-specification for Experimental and Observational data”
Jan Höffler (University of Göttingen): “ReplicationWiki: A Tool to Assemble Information on Replications and Replicability of Published Research”
Garret Christensen (BITSS): “A Manual of Best Practices for Transparent Research”
The event was organized by BITSS in partnership with the Center for Effective Global Action, the Center for Open Science, D-Lab, The Berkeley Institute for Data Science, the Alfred P. Sloan Foundation, and the Laura and John Arnold Foundation.