AEA Data and Code Guidance
Guidance for authors wishing to create data and code supplements, and for replicators.
Unofficial guidance on various topics by the AEA Data Editor
These web pages provide unofficial and developing guidance on the implementation of the American Economic Association (AEA)’s Data and Code Availability Policy. We also provide links to generic guidance being developed by a loose collective (“guild”) of data editors and people in a similar role at various social science journals.
Order in which AEA authors should read these resources:
- Start with the official Data and Code Availability Policy
- Look for general guidance at the Social Science Data Editors pages
- Read the AEA’s FAQ
- Look for any guidance specific to the AEA at the Unofficial AEA Data and Code Guidance
- Last but not least, have a look at the draft FAQ on this site
Comments are welcome, please file them as issues in our Github repo.
Guidance on creating replicable data and program archives
How should researchers create replicable data and program archives? How should such archives be structured, how documented, and where should they be located?
- Several economics and other social science data editors are collaborating on common guidance. See https://social-science-data-editors.github.io/guidance/ for more detailed guidance.
- The AEA migrated to a new data and code repository in July 2019. We provide guidance in the official Data and Code Availability Policy, the associated FAQ, additional guidance on these pages, and the draft FAQ on this site
- A DRAFT template README can be used as a starting point for a data and code archive’s master README.
- Issues about licensing are complex, we suggest the Social Science Data Editor’s discussion of licensing. The AEA Data and Code Repository defaults to a Creative Commons Attribution 4.0 International (CC-BY) license; however, we suggest that authors choose a hybrid license when their repository contains code and software, as the CC-BY license is not recommended for software.
Guidance on testing replicability of code
The code and data that have been archived should be reproducible and replicable. How do we test that?
- Generic guidance is provided at the Social Science Data Editors’ Guidance website.
- At the AEA,
- we use this template to guide our replicators.
- When we cannot access some part of the data and/or code ourselves, we may ask others to conduct a reproducibility exercise for us.
- Our PROTOCOL is outlined here.
- Sample report 1
Frequently Asked Questions
- Start by consulting the FAQ on the AEA Website. This generally apply specifically to procedures at the AEA journals.
- AEA-specific updates and additions will first appear in draft FAQ on this site
- More general questions will have answers at the Social Science Data Editors FAQ