Replicator Degrees of Freedom Allow Publication of Misleading “Failures to Replicate”

• In Event: Improving Practices of Political Analysis

Brief Overview

Commonly-exercised flexibility at the experimental design and data analysis stages of replication testing—make it easy to publish false-negative replication results, even while seeming to adhere to rigorous methodological standards.

Authors

• Christopher J. Bryan, University of Chicago

• David Yeager, University of Texas, Austin

• Joseph M O'Brien, University of Texas, Austin