Data and Code Sharing Policy

Authors of accepted manuscripts that contain empirical work, numerical simulations, or experimental work must make available to the JFE any data, programs, and details for the computations necessary for replication. As soon as possible after acceptance, authors are expected to upload their data, programs, and detailed instructions for use to Mendeley Data, where it will be published under the author’s choice of reuse license. Alternatively, authors can upload this material to a domain-specific repository. In either case, a link to the data and programs will appear on ScienceDirect with the article.

If the above requirements cannot be met, the authors should notify the editor-in-chief at the time of submission and include a precise explanation. The editor-in-chief reserves the right to refuse publication of manuscripts whose authors do not comply with these requirements.

The JFE data editor will review all code and packages to ensure they are complete and organized. Authors are strongly encouraged to use the following format for organizing their data and code packets. Details follow.

Data

Exceptions to the data-sharing policy include but are not limited to proprietary data owned by third parties whose identity can be disclosed, confidential proprietary data whose owners cannot be disclosed, confidential administrative data, experimental data that violates the confidentiality of the subjects, or hand-collected data the authors would like to keep confidential for a period not exceeding two years. In these cases, the authors are required to upload a pseudo data set. This pseudo data set should have enough observations so that all of the programs can be run successfully, including any data cleaning programs.

In the case of administrative data or third-party data whose owners can be disclosed, the authors must submit a detailed description of how the raw data were obtained or generated, including data sources and the specific date(s) on which the data were downloaded or obtained. Authors also have to provide details to outside researchers on how to obtain the data. When the data provider allows, observation identifiers in the raw data should be disclosed.

If the data are confidential and proprietary, upon acceptance of the manuscript, the authors must supply the JFE with an affidavit stating that permission has been obtained for the data to be used in the publication at hand. The JFE data editor will collect this information.

In the case of hand-collected data, if the authors would like to keep the data confidential for a period of time not exceeding two years, the authors must notify the editor-in-chief in a cover letter at the time of submission. If the manuscript is accepted, the data must be submitted to the editorial office at the time of acceptance. It will be held in confidence and be uploaded to Mendeley Data or a domain-specific repository after the embargo period.

Experiments

For experimental manuscripts, we have adopted the policies of the American Economic Review. We expect the authors of experimental manuscripts to supply the following supplementary materials:

  • The original instructions: These should be summarized as part of the discussion of experimental design in the submitted manuscript and also provided in full as an appendix at the time of submission. The instructions should be presented in a way that, together with the design summary, conveys the protocol clearly enough that the design could be replicated by a reasonably skilled experimentalist. For example, if different instructions were used for different sessions, the correspondence should be indicated.

  • Information about subject eligibility or selection: This should be summarized as part of the discussion of experimental design in the submitted manuscript.

  • Any computer programs, configuration files, or scripts used to run the experiment and/or to analyze the data: These should be summarized as appropriate in the submitted manuscript and provided in full as an appendix when the final version of a manuscript is submitted. (Data summaries, intermediate results, and advice about how to use the programs are welcome, but not required.)

  • The raw data from the experiment: These should be summarized as appropriate in the submitted manuscript and provided in full as an appendix when the final version of an accepted manuscript is submitted, with sufficient explanation to make it possible to use the submitted computer programs to replicate the data analysis.

Code

Disclosed code must include the computer programs or code used to convert the raw data into the final dataset used in the analysis. The purpose of this requirement is to facilitate replication and to help other researchers understand in detail how the raw data were processed, the final sample was formed, the variables were defined, any outliers were treated, etc. Also required are any programs used to generate the final published output from the final dataset.

All of this code must be accompanied by instructions for use. These instructions should include information on the operating system and the version of the software used in the analysis. Software requirements should explicitly list all modules, libraries, toolboxes, packages, and commands that are not part of the core software. Programs that rely on random number generators should set seeds to allow replication. If it is necessary to execute programs in a particular order, these details must be included in the instructions.

Any person downloading any of the file(s) will need to certify that the programs will be used only for academic research. Any other use, including for commercial purposes, is strictly prohibited except with explicit permission from all authors of the published article. If the code comes from a third-party source, explicit permission must be obtained from the cited originators of the code before use. Academic researchers using the code, or fragments of it, in their own work are required to acknowledge the origin of the code.

Exceptions to the code-sharing policy may be granted if the authors invested a great deal of intellectual effort in generating the code and want to keep it private for a period not exceeding two years. In this case, the authors must notify the editor-in-chief in a cover letter at the time of initial submission, and the code must be submitted to the editorial office at the time of acceptance. It will be held in confidence and be uploaded to Mendeley Data or a domain-specific repository after the embargo period.

Exceptions to the code-sharing policy will also be granted if all or part of the code is proprietary in nature or if parts of the code reveal confidential information.

Post-Publication Author Responsibilities

Authors are not required or expected in any way to provide assistance and further support to eventual users of the code and data.