FAIR and open practices in archaeological research: Salmo sp. open code publication
- Melissa Kays
- Ben Catt
Dr Liz M. Quinlan, Department of Archaeology
Email: liz.quinlan@york.ac.uk
Summary
This case study engages with open and FAIR (Findable, Accessible, Interoperable and Re-usable) data practices in the context of archaeological research tracking the decline of salmon in the North Sea basin. The process of documenting and organising R code for public access has helped Liz to enhance her understanding of FAIR principles and develop new skills and approaches in her work. By sharing her data and code openly she has enabled others to understand, reproduce and adapt her research, whilst encouraging them to engage with open data practices and FAIR principles in their own work.
Case Study
This open data project is a key part of the PhD thesis: ‘Medieval icthyoarchaeology and historical ecology of Salmo sp. fishes across the North Sea Basin’. This thesis work involved the analysis of archaeological fish bone assemblages, documentary records, bulk stable isotope data, and proteomics-based species identifications– all of which required statistical analysis and data visualisation to aid in interpretative conclusions.
As part of a Horizon 2020 funded doctoral research project, Liz agreed to uphold the open science objectives of the European Commission, which centre around the generation of open access and FAIR data. With the aim of both complying with these requirements and contributing to ongoing efforts to standardise open data practices, all code produced as part of this PhD work will be published as both an appendix to the main thesis and as a publicly accessible GitHub repository. In order to not only fulfill this objective, but also record progress and enhance data analysis and visualisation skills, Liz began compiling an rMarkdown file of all code written in R for the thesis.
Prior to beginning the PhD in October 2019, Liz had no experience in coding using the R programming language, nor with the compilation of open data resources. The process of annotating, recording, and organising statistical and visualisation code to be shared publicly allowed Liz to gain a deeper understanding of how FAIR practices need to be implemented just as much in small personal projects as in large research initiatives.
As someone new to programmatic statistical analysis and data visualisation, Liz struggled to find resources on using R for archaeological data analysis that fully explained the decision making process in a way that was accessible to novice coders. Even in publications with “open access”, “open science” and/or “open data'' commitments, code was rarely included, and when included it was the “best version”, with minimal documentation and limited reproducibility. Liz intentionally organised the rMarkdown thesis code document chapter-by-chapter, with the ultimate intention of making all associated isotopic, proteomic, and zooarchaeological review data available online. This will enable anyone to download the data and rMarkdown file, and run all code for the project on her own to reproduce the research in its entirety and investigate other avenues of analysis as desired.
The intention is for this to not only serve as a resource for other early career archaeologists, but to start a conversation about individual responsibility for implementing FAIR principles. Engaging with open data practices throughout the PhD experience has not only allowed Liz to understand more about the need for consistently applied– but still situationally appropriate–standards of data sharing, but also improved her personal research skills. Writing code with the understanding that someone else will need to understand, reproduce, and adapt this work has made Liz more mindful of her analytical choices, and fuelled a desire to encourage others to engage with open data practices in the future.
Links
Acknowledgements
Funded by: Marie Skłodowska-Curie grant agreement No. 813383 - SeaChanges Innovative Training Network
Licensing information
Except where otherwise noted copyright in this work belongs to the author(s), licensed under a Creative Commons Attribution-NonCommercial 4.0 International Licence