Getting research data ready for peer review

Sharing data underlying articles substantiates your research and allows others to use data. There is a movement towards underlying data always being shared. With that in mind, and for Peer Review Week 2022, here are our tips on how to get your research data ready for peer review.
Published in Research Data
Getting research data ready for peer review
Like

Do you have to share your data during peer review?

Yes, if requested to do so. Even if your data cannot be made publicly available upon article publication, the Editor and peer reviewers should be able to properly review the data alongside the manuscript.

Furthermore, if there will be any restrictions on data availability once your article is published, these “must be discussed with the editor at submission who reserves the right to decline the study if these conditions are found to be unduly prohibitive” (from Nature editorial policies).

Should you publish data prior to submitting your article?

Publishing your data prior to your article has advantages. Firstly, it makes it easier to share your data with peer reviewers. Secondly, in a similar manner to pre-prints, sharing data can act as a proof of work done. With this in mind, you may consider a data publication. At Springer Nature, we have two kinds of data publication: Data Descriptors, which are highly detailed, data-focussed papers that link to the published data, and Data notes, which are briefer and with an emphasis on speedy peer review. 

How can data be made available to peer reviewers?

First, check what data repositories serve your research community. It may be that these repositories facilitate private peer review (i.e., can make the data available via a private URL prior to data being published publicly). Note that for certain types of data, there are mandates requiring that the data be shared in certain repositories. Use this table of mandated data types to check.

Here are a few tips to find appropriate data repositories:

  • Check recent publications in your field to see if they have shared their data in a repository, or if they cite/mention other data.
  • Use these registries to search across a large number of repositories, filtering by various categories: https://www.re3data.org/ or https://fairsharing.org/.

If no specialist repository exists for your type of data, or if the specialist repositories do not support private peer review of data, and you are choosing not to publish your data prior to your article, then you can use a generalist repository which supports private peer review. Here are some that do: figshare, Science Data Bank, Zenodo, Dryad, Open Science Framework, Harvard Dataverse.

If using a generalist repository, bear in mind that reviewers/editors may request upload to a specialist repository upon manuscript acceptance.

How should you present your data for peer review?

Data should be presented in a way they can be understood and analysed. This means:

  • Complete, with respect to the manuscript, so that all data underlying the outcomes of the manuscript are included. Note: the summary statistics (means, standard deviations, population sizes, etc) are not sufficient--it is the data underlying the summary statistics that should be available.
  • In an open format (.txt, .csv, .xlsx, etc), as far as possible. If this is not possible, then the software required should be declared (the manuscript’s Data Availability Statement should contain this information, and you may also include it in the READ ME).
  • With appropriate embedded metadata. For example, spreadsheets should have understandable column headers.
  • In English (or in the language of the journal your manuscript is under peer review at, if not English). This applies to the headers within the data files, any other embedded metadata, and the READ ME. It can also be helpful for file names.
  • If the data require code to be opened or analysed, this code should be either linked to or included with the data, and a description of how to run the code (i.e., a READ ME file) should be included.

How can you let peer reviewers know how to access your data?

If your data are already public during peer review, you can just add the data DOI to the Data Availability Section, along with a brief description of the data (name of repository, number of files, access conditions, etc). If your data will remain private throughout peer review, you can use the same method, but you must take care to either 1) publish the data and replace private/peer-review links with the data DOI or accession code after your manuscript is accepted, or 2) remove the private link from the Data Availability Statement prior to manuscript publication (this is not an option if your data are of a mandated type).

If you have other questions about preparing your data for peer review, or other research data-related questions, feel free to contact Springer Nature’s Research Data help desk.

*banner image for Peer Review Week 2022 from this page: https://peerreviewweek.wordpress.com/ 

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in