Practical steps toward more reproducible research

The Open Research at Cambridge conference took place between 22–26 November 2021. In a series of talks, panel discussions and interactive Q&A sessions, researchers, publishers, and other stakeholders explored how Cambridge can make the most of the opportunities offered by open research. This blog is part of a series summarising each event. 

On 26 November 2021 the University’s Reproducibility Working Group hosted a workshop for researchers from across Cambridge to explore approaches to supporting more reproducible research. Talks were provided by Professor Alexander Bird (Faculty of Philosophy), Dr Florian Markowetz (Cancer Research UK Cambridge Institute) and Dr Maria Tsapali (Faculty of Education) exploring approaches to reproducible research and reasons to work reproducibility across qualitative and quantitative research.

The recording of the session can be found below:

Talks were followed by interdisciplinary discussion sessions designed to identify the obstacles to reproducible research across Cambridge and how these might be tackled.  The key findings from the discussions included:

  • Training on reproducibility, including statistical training, reproducible methods and use of key tools exist in departments across the University, but more needs to be done to share provision and create synergies and central provision where possible. 
  • Training should begin at undergraduate or Masters level to build key skills early.
  • Awareness of training, and the importance of reproducibility training, needs to be enhanced.
  • The need for University guidance on how to make research reproducible, particularly to overcome key challenges to reproducibility such as balancing reproducibility with the need to protect sensitive or confidential data.
  • That the University can help by making the production of open and reproducible research as painless as possible, for example by facilitating peer review of codes and providing easy access to data storage and expertise in best practice.
  • That reproducibility looks very different across the disciplines and that in some areas transparency and methods reproducibility will be the focus, rather than reproducible outcomes.

The Reproducibility Working Group will draw on the ideas raised at this workshop to help shape proposals for future University approaches to supporting reproducible research. The group plans to host a number of further events to map, consolidate, and extend existing resources for reproducibility across Cambridge with the aim of boosting grassroots activities and magnifying their impact across all levels of the institution.

For more information and resources on reproducible research see: UK Reproducibility Network: https://www.ukrn.org/

Open Research in the Arts and Humanities – beyond Open Access

The Open Research at Cambridge conference took place between 22–26 November 2021. In a series of talks, panel discussions and interactive Q&A sessions, researchers, publishers, and other stakeholders explored how Cambridge can make the most of the opportunities offered by open research. This blog is part of a series summarising each event. 

A conversation with Emma Gilby and Rachel Leow, chaired by Matthias Ammon

This session was based on the work of the University’s Working Group on Open Research in the Humanities. The main activity of the group, which was formed in summer 2021 and is chaired by Emma Gilby, Professor of Early Modern French Literature and Thought in the Faculty of Modern and Medieval Languages and Linguistics, has been to discuss ways to make some of the underlying principles of Open Research – which have often been based on scholarly communication in STEMM subjects (as for example defined by the League of European Research Universities as the ‘8 Pillars of Open Science’) – more applicable to humanities research. The conversation was also intended – as implied in the title – to explore issues around humanities research that went beyond the mechanisms of Open Access publishing, to consider the production of research as well as conditions of dissemination.

Some of the discussion about applying the concepts of Open Research to the humanities centred around differences between research in the sciences and the humanities. For example, the concept of ‘reproducibility’ may not necessary be a quality that’s applicable to humanities research, which does not aim to produce reproducible results via experiment, but builds on and recontextualises earlier discoveries, methods and debates. Research integrity here comes with a vital element of care – care to represent source material fairly but also awareness of the scholarly ecosystem, the processes behind scholarship and the networks of people involved. For instance, publishing in the humanities in particular relies on the labour of editors, copy editors and typesetters, among others, all of whom share (along with the author) care and responsibility for disciplinary values and standards, and whose work needs to be recognised and acknowledged.

The concepts of care and openness also appeared again in a discussion of what is often considered one of the major benefits of Open Research, namely that it makes scholarship more globally accessible. Careful consideration here needs to be given to the issue of appropriate and fair representation across a diversity of voices and communities. For instance, digital archives tend to reproduce English-speaking structures and skew towards information that is already easy to find.

These are a couple of specific examples which demonstrate that the underlying structures of the way research and scholarly communication are conducted in the humanities require a significant amount of rethinking of the concepts behind Open Research in a humanities context. The Working Group is currently producing a report which will discuss these and other aspects of Open Research in more detail and make suggestions for how institutions such as libraries can support researchers in this context.