As an editor of a journal with a double-anonymous peer review system, I often wonder about the right balance between open science practices and anonymisation of the manuscript for the review process. How much anonymisation is enough while being compatible with open science dissemination?
In particular, when a manuscript includes information about the protocol registration and raw data, should we ask the authors to anonymise this information as well? And what is the proper balance between open science practices and maintaining the integrity of the single or double-blind anonymous review process?
Questions for the Forum
- When a manuscript sent for anonymous or double-anonymous peer review process includes information about the protocol registration and raw data, should the editor ask the authors to anonymise this information as well?
- Does the Forum have any other advice or comments on balancing these competing demands?
In science generally, and in medicine particularly, the protocol information forms a vital part of the scholarly process and should be available for reviewers to assess. It is not uncommon for reviewers to find manipulation or errors in the data, and if they are not shared then serious publication ethics issues may be missed. In some cases, it may be possible to anonymise these data, but editors may also be able to call on in-house experts to examine these files separately from the main review process. This would mean that the information need not necessarily be anonymised. However, the key fact is that assessment of protocol data is sufficiently important to outweigh a policy of anonymisation.