Publication Ethics Resources

creating publication standards

This is a public AAA Task Force  public

Discussion

    Julie Smith David
    Updated Draft: Data Policy as of 10/28/2014
    Discussion posted October 28, 2014 by Julie Smith David, last edited July 24, 2015, tagged completed 
    520 Views, 3 Comments
    Title:
    Updated Draft: Data Policy as of 10/28/2014
    More:

    Based upon feedback from the Annual Meeting, the Publications Committee and Research Committee, the PETF has updated the draft data policy.  Please comment below or email Mary Stone at mstoneATcba.ua.edu if you have suggestions for improvements.

    Comment

     

    • Shailendra Pandit

      I like the idea of having a clear data policy for AAA publications. Researchers should be responsible for ensuring that a third party can replicate their results, particularly when publicly available data sources are used in the study. To that end, making programs/code available is a good first step. My co-authors and I have previously been requested by a (non-AAA) journal to make our programs available for posting on the journal website, and we happily complied.

      However, there could be ownership/copyright issues involved in making data files available to third parties. In our conversations with a leading vendor of firm-level financial data, my co-authors and I were told by the vendor representative that making our programs/code available was fine with them (but possibly subject to copyright rules, if any, of SAS/Stata, etc.). But the vendor rep took issue with making our data available to the public at large stating it would violate our data use agreement with the vendor. It seems that one of their main concerns is unauthorized access to subscription-based data by non subscribers.

      Therefore, my recommendation for the draft data policy is to consider the language of recommendation carefully so that it does not place unnecessary burden on researchers that could place them in conflict with data providers.

    • Linda Achey Kidwell

      I was just having a discussion with colleagues in finance about data and replication.  For some of their journals, they are required to maintain a step-by-step description of their selection of data for archival research with market data, such that somebody else could follow their list and extract exactly the same data. One implication, for example, is a detailed description of why certain firms or observations were thrown out of the set.  They also must maintain their records of the judgments they made when coding international data so that it could be replicated.  Perhaps the AAA should require such descriptions be submitted (to the editor, not necessarily for review) and retained until some period after publication.  I would also say a good preventative measure for archival study integrity is to begin publishing replications, as many other fields do.  That might discourage data falsification for fear of discovery, which without replication is highly improbable.

      For behavioral research, of course replication is inherently more difficult, but a policy of instrument and data retention for a period of 5 years would be advisable.  However, the policy must not run the risk of violating IRB guidelines regarding protection of subject rights.

    • Julie Smith David

      “I like the changes made to this draft policy. It is clear that the committee heard the comments made previously and was open to incorporating them into the revisions. I am particularly gratified that the role of the Editor, as central to this process, has been noted. However, I would like clarification on the Editor’s role in the event that a “special committee with representation from legal counsel be appointed to determine additional steps and/or corrective action(s) to be taken.” Will the Editor be a participant on this committee?

       

      posted on behalf of Mary Curtis, Senior Editor, Journal of Information Systems