Review of How fair is bioarchaeological data: with particular emphasis on making archaeological science data reusable
For Peer Community in Archaeology, by Jan Kolar (UCL Institute of Archaeology)
 
The paper aims to evaluate bioarchaeology from the perspective of data accessibility and reusability. The author conducted a needs analysis for identifying existing practices and needs of the bioarchaeological community.
His main conclusions are:
-       Bioarchaeology significantly reuses data from other papers.
-       There is no standard process of data deposition and back-up across the subdisciplines of bioarchaeology.
-       The bioarchaeological community perceives the importance to store and share the data transparently, but only part of it is doing so.
 
The paper certainly aims to discuss important issues of current scientific practice – fairness and open access. The author decided to analyse that through an online survey and then discussing the results. The intention behind this endeavour is good, the paper fails to provide an up-to-date and broad analysis. Firstly, there are no references dated after 2020 and even from that year there are only couple of them. The author does not even refer to an extensive review of data sharing in archaeology (Marwick and Birch, 2018). There are no mentions to open access aDNA databases and databases of isotopic datasets (see my comments below), so it seems that the author is not aware of their existence. Secondly, the paper is highly over-visualised (31 figures!) which makes the text very fragmented and not easy to follow. Thirdly, there are no clear conclusion in the Conclusion section.
 
This paper needs a lot of work and I cannot recommend its publication in the current state. However, I believe that after substantial amendments it can be published in a form of a short report with much lower number of figures. That of course depends on the place of submission.
 
Major Issues:
Through the paper – The author refers to the paper as the “thesis”. Is it because it was originally part of a thesis? Confusing, please amend.
Line 79 and further: Background: This section does not belong to Materials and methods as it explains the background concepts of the paper and at the end it presents the main aims of the paper. The whole section should be moved to the introduction, just after the existing text there. Then it would have a logical order.
Lines 125-127: Here the author finally claims that the paper will deal with the biomolecular archaeology. The focus of the paper was so far slightly confusing, as the author was all the time speaking about bioarchaeology as whole, but still not mentioning e.g. archaeobotany. The focus should be clear from the beginning, so please narrow your subject of interest much earlier, or keep it wide for the whole paper. Be consistent.
Line 128 and further: Ancient DNA section: Here the author somehow does mention only the need to store the raw data of aDNA without mentioning that it is actually happening. I recommend checking Orlando et al. (2021), which has references to open access aDNA databases.
Line 142 and further: Stable isotopes section: here the author again presents what are stable isotope analyses but does not provide any references to open access repositories of these analyses (e.g. Plomp et al., 2022). I believe that is missing as open access repositories are the main topic of this paper. Please amend.
Results: The author uses throughout this section pie charts for visualisation of the results of the survey. Although pie charts are used relatively frequently in archaeology (I am also guilty of that crime on data visualisation), pie chart is not the best method for such visualisations. Bar charts and histograms are much better. The author can find the details and instructions for better visualisations on the links below:
https://scc.ms.unimelb.edu.au/resources/data-visualisation-and-exploration/no_pie-charts
https://medium.com/analytics-vidhya/dont-use-pie-charts-in-data-analysis-6c005723e657
https://www.data-to-viz.com/caveat/pie.html
line 211, Findability: Here the author discusses use of doi and ORCID. In case of doi he does not mention them, and speaks only generally about persistent identifiers. It is not clear why. To me it is also not clear why does he speaks in this paper about ORCID, which is used to identify authors and link them properly with their published papers. How is this related to the use of persistent identifiers for datasets is not clear from the text. Please clarify.
Results: Overall, this section is too fragmented and many sections have just few sentences and then many figures. This paper seems to be “overvisualised”. It is certainly desirable to have good diagrams in our papers, but 31 figures in this paper is just too much. I don’t know where this paper will be published, but the usual limit in scientific journal varies between 5 and 10 figures per article. I strongly recommend cutting the number of figures to max 10 (the most important ones) and build the paper around them. This is anyway the maximum a reader can comprehend.
Line 225: here the author claims that 40% is majority. Majority of a whole is more than 50%, 40% in this case is the largest proportion.
Figures 15 and 28: These figures are totally ineffective and not understandable. I recommend another visualisation which would be easily comprehensible.
Figure 23: The barplot is useless, the table under is sufficiently showing the data.
Figure 24: Too many categories, too complicated, not clear what is the actual difference.
Line 288: “Possibly one of the most restricting elements is the lack of Data Management Plans, with over three-fifths of the respondents stating they do not create them.” Restricting in what sense? Unclear. Please clarify.
Line 296 and further: Analysis. Structurally, analysis should be under the results section. Please omit this heading and simply add this summarizing part at the end of your results section.
Lines 314-318: As I outlined already above, the author mixed up doi (for datasets) and ORCID (for authors). This is confusing and the use of ORCID by the authors of datasets cannot be used as an argument for findability of datasets as ORCID is not used for identifying datasets.
Lines 319-322: The statements regarding datasets accessibility in this section are contradictory. Please clarify.
Figure 31: This figure (technically a table) is not fully clear. What does DMP stand for? Percent of what? Please clarify.
Line 363-378: 7.b Again the already mentioned issue with mixing up doi and ORCID. Please clarify.
 
Minor Issues:
Line 33-34: “understanding communities” – It is not clear what is meant be that. Please clarify.
Line 38-39: “damaging the raw material” – Does the author mean by that destroying the original samples? I think “raw material” is not appropriate in this context.
Line 57: “Methods and materials” – This section should be rather called “Materials and methods” as is more usual in research papers.
Line 58: “Methods and Materials” – duplicated heading within the paragraph. Please delete.
Lines 323-237: Would be interesting to analyse also demographic of the responders? DNA specialists tend to be younger, more familiar with databases etc, tend to receive funding which requires open data and tend to publish in journals which require repositories.
Line 339: Is this a dissertation or a paper? Please clarify and refer correctly.
Line 429: Is this a thesis or a paper? Please clarify and refer correctly.
 
References:
Marwick, B., Birch, S.E.P., 2018. A Standard for the Scholarly Citation of Archaeological Data as an Incentive to Data Sharing. Adv. Archaeol. Pract. 6, 125–143. https://doi.org/10.1017/aap.2018.3
Orlando, L., Allaby, R., Skoglund, P., Der Sarkissian, C., Stockhammer, P.W., Ávila-Arcos, M.C., Fu, Q., Krause, J., Willerslev, E., Stone, A.C., Warinner, C., 2021. Ancient DNA analysis. Nat. Rev. Methods Primer 1, 14. https://doi.org/10.1038/s43586-020-00011-0
Plomp, E., Stantis, C., James, H.F., Cheung, C., Snoeck, C., Kootker, L., Kharobi, A., Borges, C., Moreiras Reynaga, D.K., Pospieszny, Ł., Fulminante, F., Stevens, R., Alaica, A.K., Becker, A., De Rochefort, X., Salesse, K., 2022. The IsoArcH initiative: Working towards an open and collaborative isotope data culture in bioarchaeology. Data Brief 45, 108595. https://doi.org/10.1016/j.dib.2022.108595
 

The manuscript submitted has an interesting approach regarding open access on bioarchaeological data and their opportunities to reuse and maximize the benefits derived from archaeological research. It presents a coherent structure with an introduction to the main disciplines that study bioarchaeological material (with the exception of archaeobotany, which is not mentioned in the manuscript). Secondly, it presents methods and materials, where the authors used A Need Analysis based on questionnaires answered by email. Although it is spread in different countries and institutions, it comes to the attention that only includes the UK, some of the Institutes in the United States, and some institutions from other parts of Europe. Although it is a wide spectrum, I considered that some institutions are not represented (e.g. Latin American, which in some cases has a strong position regarding open access archaeological data). 

As for the results presented in the manuscript, they are coherent with the methodology planned and the research questions. Discussion and interpretation of these results allows the authors to attend to the different parts of the preserving, sharing and reusability of the data.

I fully support the overall message of this article, which is advocating for the use of the FAIR data principles in Bioarchaeology, however, I think it needs considerable reworking and revisions to make it of publishable standard.  

• The paper would benefit from being restructured – you have background sections after the materials and methods section. The background sections on FAIR and the intros to the different types of bioarchaeological data should be part of the introduction. Analysis section should be part of the discussion section. 
• Methods section needs much more development - your methods of data collection, data processing and analysis need to be described in much more detail and your article would benefit from additional documentation around the methods such as a questionnaire document that showed all the questions.
• You are lacking many references relating to general archaeological, zooarchaeological and archaeobotanical data management.
• I also don’t think that your data does support your analysis – I think you have overstated how well bioarchaeologists are making their data reusable.