Assistant Director, Research Data and Metrics/Vice Chair for Research NYU Health Sciences Library New York, New York
Background: In 2016, our library was invited to teach a required 1-credit research skills class for first year PhD students in biomedical sciences, which included lessons on rigor and reproducibility, literature searching, data visualization, research data management, and other topics. In response to changing NIH requirements on rigor and reproducibility education for training grant recipients, we updated our curriculum for the 2019-20 academic year to be a focused course on Rigor and Reproducibility rather than a general skills class. Briefly, rigor is the scientific soundness of research, and reproducibility is if an experiment can be repeated with similar outcomes.
Description: Our team of librarians took relevant elements from our previous class and reframed them around Rigor and Reproducibility, as defined by the NIH and National Academies of Science. We also created new educational content focused on replicability (the ability to repeat a study with similar results) and computational reproducibilty (the ability to re-run analyses), with an invited speaker discussing authentication of biological resources. The class is a mixture of lecture, hands on activities and technology-facilitated discussion. Our goal was to provide a unified educational class to better meet learner needs, and student performance is assessed by summative and formative assignments, as well as in an attitudinal survey at the class's end. We compared the updated class with past years to understand if a focused class on Rigor and Reproducibility is better received than a multi-topic series.
Results: A majority of over 2/3 of students found the new rigor and replicability focused material moderately or very useful and indicated by their own self-assessment an increase in knowledge on class topics after the class. All students passed our final summative evaluation, and instructors reported feeling that the new class had fewer issues and students demonstrated a greater capacity to grapple with topics as compared to the old class offering.
Conclusion: While valid year-over-year comparisons between different classes are difficult to demonstrate by scores or evaluations, the new iteration of the course was well reviewed in terms of students' self-evaluation of learning and utility. Moreover, while the final exam changed, instructors felt learning was more apparent in the new Rigor and Reproducibility class. Instructors also felt they had a superior experience teaching the new Rigor and Reproducibility curriculum as compared to the old skills-based-class.