This document describes how to best prepare a replication package for an AEA journal. Much of the guidance here is not specific to our journals - the document links to other websites for tutorials, best practices, etc. The best moment to do the preparation described here is, in fact, when you start the project, not once you have had your manuscript conditionally accepted. However, all steps here can and have been successfully performed at the point of conditional acceptance.
The Data Editor of the American Economic Association endorses DCAS, the Data and Code Availability Standard v1.0. The AEA’s data and code availability policy is compatible with DCAS.
Assistance¶
If, in the process of preparing your replication package, you have questions that are not addressed by these guidelines, please contact us:
Contact us via the AEA’s Contact form
Reach out to the editor handling your paper, who knows how to reach us (but might also have the answer).
Reach out to us directly (you know how to find us).
Key principle: Computational empathy¶
Keep in mind: The replication package is meant to be run by others, who have none of the setup, packages, and data that the original author might have, on computers that may not run the same operating system.
Treat the replication package as one of the methods to convey the methods that lead to your manuscript’s conclusions. Consider it a teaching tool, targeting young graduate students who may not be in your field.