Lay summary authored by Lisa Lodwick. The full article can be read in Open Quaternary: http://doi.org/10.5334/oq.62
Archaeobotanists spend countless hours at the microscope, sorting and identifying ancient plant remains. Counting seeds is a key step in the research process, enabling patterns to be established across sites, within regions, and through time – patterns that can tell us about a wide range of past human-plant relationships. As the number of archaeobotanical data sets have grown and statistical analysis has become more widely used, meta-studies have provided new insights into the timing of crop domestication, the spread of domesticated plants, and shifts in how crops were cultivated. Undertaking such meta-analysis relies on access to the counts of plant remains from the initial studies. However, there is a common perception that much data produced within archaeology, and across many academic disciplines, is not always made publicly available. How much seed count data is still available?
In this study, I used over 200 journal articles published over the last decade in 16 journals to assess how archaeobotanists are currently sharing their data. Whilst 56% of papers did provide their full or ‘raw’ data, this was often in forms that make the reuse of this data challenging – e.g., the data might need re-entering into a spreadsheet. The remaining articles did not contain their ‘raw’ data, limiting the types of analysis for which they could be used. A second assessment was undertaken of meta-studies using previously published archaeobotanical data – whilst 64% did reference the studies used, 21% did not, hindering the extent to which others can build on these studies in the future. The lack of formal credit, by way of citation, to data producers for the creation of archaeobotanical datasets may limit the motivation to make future datasets available in a reusable format.
The decisions researchers make regarding if and how to share datasets is complex, with factors including career stage, access to financial and technological resources, education and training all coming into play. But discussions of data-sharing should be undertaken across academic communities, and training should be provided as part of degrees and within professional bodies; after all, the availability of sample-level data enables the fullest future use of archaeobotanical research, and makes the most of long hours spent at the microscope.
Based on the article:
Lodwick, L., 2019. Sowing the Seeds of Future Research: Data Sharing, Citation and Reuse in Archaeobotany. Open Quaternary, 5(1), p.7. DOI: http://doi.org/10.5334/oq.62