Fears of a “reproducibility crisis” have led scientists, funding sources, and the general public to question both the efficacy and trustworthiness of scientific research, especially in biomedical, psychological, and social sciences. To date, discussions have focused on policy changes and behavioral regulation related to statistical practice and experimental design. However, “reproducibility” is a broad concept that includes a number of issues, many of which relate to theoretical assumptions and inferential methods that are unaddressed in many current initiatives. The mixed results of explicit endeavors to reproduce past research highlight that interpreting the significance of replication experiments is difficult. Additionally, some types of reproducibility failure are a normal part of genuine inquiry that scientists use strategically to produce more integrated explanatory frameworks.
The Many Faces of Reproducibility is a three-year project supported by the College of Liberal Arts at the University of Minnesota involving an interdisciplinary exploration of these different facets of reproducibility across the social, behavioral, and natural sciences. It aims to advance scholarship on reproducibility, develop innovative approaches for training researchers, and engage the public regarding the trustworthiness of scientific claims.