A practical approach to teaching reproducibility, and improving your own research at the same time.
Moderated by: Michael O'Hara
Assistant Professor of Economics at St. Lawrence University
In the present talk, I describe my teaching approach for a 300-level undergraduate-only methods and statistics course in psychology. This course centers reproducibility, and the main learning goal is for students to be able to identify and create reproducible research projects. As part of my approach, I identify and teach three main components of reproducibility. The first is prose-based reproducibility: based on the content of a Method section, is it possible to reproduce the methods? The second is analytic reproducibility: based on the contents of an OSF page, a preregistration, and a manuscript, is it possible to reproduce the analyses reported in the manuscript? The third is what I call "design reproducibility": based on the previous literature, is it possible to reproduce the thought processes and decision-making processes that led to the methodological and analytic choices made in the study? I argue that these three elements of reproducibility are essential for high-quality research, and should be systematically trained starting at the undergraduate level. In my talk, I address specific assignments that help students to answer the three questions described above while simultaneously training them in the skills required to create reproducible work (e.g., RMarkdown, preregistration writing, OSF organization, effective Method section writing). The final projects for the course involve writing up an empirical experiment (with reproducible analyses and records available to my and other students) and developing a truly reproducible preregistration. A secondary focus of my talk will be on the practicalities of implementing such assignments into courses, and, more generally, of centering reproducibility in a moment when many of us lack for time and resources.
Dr. Sullivan is an Associate Professor at Skidmore College. Dr. Sullivan studies how children learn about -- and from -- the world. Her specialties are in language, numerical, and social development. Dr. Sullivan has created alien languages in order to study how children learn new words, and has tested the impact of a child's language on your beliefs about math and infinity by testing children who speak different languages (Slovenian, Cantonese, English, French, Spanish, Hindi, Saudi Arabic, and Gujarati) on the same task. Dr. Sullivan also focuses her research on social development. Some of her studies have tracked how gender stereotypes change across the lifespan, and have shown that preschoolers who violate gender stereotypes sometimes experience penalties for doing so. Most recently, her work has focused on understanding how and why adults talk with children about race.