Archive, or not to archive?
21 February 2013

Immediately after MRS CEO Jane Frost circulated a note to members announcing MRS new initiatives in January, including the launch of the new IJMR website, I received an email from an Editorial Advisory Board member asking if the opportunity had been taken to include a database facility to enable author’s datasets to be archived for access by IJMR readers.

I had to reply that in fact this possibility had not even crossed our minds, but promised to raise it at the January Executive Board meeting.

Whilst, for example, the Royal Statistical Society requires datasets to be submitted wherever possible in support of submitted papers, IJMR has never had a policy of requiring this from authors. 

However, further research on the topic of archiving revealed a current lively debate in the field of medical science on a website called ‘The Scholarly Kitchen’ led by Tim Vines, which you might like to look at. Vines is the joint author of a paper on this topic, referred to in his post, which concludes that a mandatory policy is the only effective way to achieve this goal, and that journals should seriously consider adopting this policy, if only to ensure that datasets remain in existence.

As promised, we duly discussed the issue at the Executive Board meeting, and decided after much debate not to introduce such a requirement. So, why did we make this decision? 

• Firstly, this would require very complex changes to the MRS website – not just to hold datasets, but to facilitate access or analysis by readers. 

• Secondly, at IJMR we encourage submissions from practitioners, who are often reporting on client specific data with confidentiality issues, and we felt this would lead to even fewer such submissions – which are already declining as a proportion of all submissions. 
• Thirdly, we felt that this posed issues for submissions from other countries, for example, the possible need for translation into English.
• Fourthly, referees are free to ask authors for supporting evidence if it is felt there are concerns about the basis of claims made in a paper – it is all about trust. 

Many submissions are already rejected prior to peer review due to concerns about the methodological basis, which authors are unable to satisfactorily address. Often in judging the worth of a submission, the questionnaire can be of more importance, or the sampling and data collection methods. 

As has been widely recognised, academic papers often focus on the literature review, developing hypotheses, analysis and implications for managers, with scant attention being paid to the research design. Submissions to IJMR are no exception – but the survey methodology, if that’s the basis of the paper, is vitally important for IJMR. 

However, we recognise that in some fields of scientific endeavour, especially areas such as medicine, this should be mandatory, and this is likely to be of increasing importance if open-access becomes the requirement where research is funded from the public purse.

Do you think we’ve made the right decision? Your views would be very welcome, including your experience of other journals.