Tag Archives: unibo

W-log #4: events, reflections, and readings about reproducibility

What happened? A few days ago, the University of Bologna organised a half-day event entirely dedicated to the concepts of reliability, transparency and reproducibility of research. It counted international experts, including Sabina Leonelli and Daniele Fanelli, who gave two exciting keynote speeches, and two roundtables with several institutional representatives (from the National Agency for The Evaluation of Universities and Research Institutes, the Ministry of University and Research, the UK Reproducibility Network, and the Jagellonian University) and with UNIBO’s researchers (from PhD students to Professors) introducing their take on reproducibility within their research and discipline. It has been a wonderful moment of discussion within our community, which will be followed by two additional events (in October and December) dedicated, respectively, to interdisciplinarity and ethics.

What I did. As a consequence of my take-home messages from the event and the slides I prepared for a lecture I gave to second-year students of the Cultural Heritage in the Digital Ecosystem PhD programme, I spent several hours on the concept of transparency, which is strongly tied to reproducibility. In several meetings and papers, I’ve seen too many times that transparency is introduced as synonymous with quality. While I understand the rationale behind such a conclusion, I believe this claim is rather simplistic – and wrong, from my perspective. I would need to spend more effort (and words) to rigorously outline the whole rationale of my thoughts – at least more than a short paragraph in a blog post. Still, the main point here is that quality is a goal to achieve (e.g. to perform qualitatively sound work), while transparency is, in fact, a tool that enables others to check if we have reached that goal. In practice, quality can exist without transparency, but transparency is needed if we want people to scrutinise quality. Thus, transparency does not build quality; it builds trust.

What to read. With all this agitation about these topics I’ve been absorbed this week, the weekly suggestion is, indeed, about research reproducibility and shows how there is not one definition that fits all but, rather, a plethora of different lenses one may have to use, depending on contextual and/or disciplinary situations:

Goodman, S. N., Fanelli, D., & Ioannidis, J. P. A. (2016). What does research reproducibility mean? Science Translational Medicine, 8(341). https://doi.org/10.1126/scitranslmed.aaf5027 – Open Access at https://osf.io/dw23g