Data and Reproducibility

The publisher encourages authors to share and report meta-data, code, and materials for the transparency, replicability, and trust in scientific findings.

Data sharing

Authors are encouraged to submit meta-data with their manuscripts or to preserve the datasets on a certifiable disciplinary data repository, which ensures that editors and reviewers can access to the datasets. The links to publicly accessible datasets should be available, which are generated during the study. If the datasets are available publicly, authors should declare the data availability statement in the manuscript, which is quite valuable for further reproducibility. Exceptions exist if the datasets involve confidential data, when authors must explain it to the editor in a cover letter.

To prepare a data available statement:

  • If the datasets are preserved in a data repository, persistent identifiers (e.g., DOIs) should be provided for access. Authors are recommended to follow the FAIR Data Principles, which suggest authors to choose a data repository with persistent identifiers, and to make research data “findable, accessible, interoperable, and reusable”.
  • Authors should provide any original data and secondary reuse data, which support the analysis and results of the article.
  • Authors should explain the reason for not sharing the data if datasets involve privacy or confidential data referring to the discipline guidelines.

 

The journal editors and reviewers are responsible for the data reliability, and a data quality-reviewed process will be conducted; the journal will address concerns to unpublished datasets according to the guidelines of Unpublished data flowchart of COPE. If there are data errors, incomplete datasets, or data manipulation, journal editors will contact the corresponding author for responses to the concerns. Necessarily, the journal office may cooperate with authors’ institute to investigate the whole study. Supporting documentation is needed. If the authors’ responses do not give a satisfactory update to the dataset, or the issues are too severe, the authors should withdraw the data deposition from the repository, and the journal will reject the manuscript.

If a manuscript reuses published datasets whose scientific rigor is in doubt, the corresponding editor and implicated journal will be contacted. The journal office will address the concerns according to the guidelines of Published data flowchart of COPE. A correction may be posted, and the latest dataset version with its updated DOI is linked to.

 

Data citing

Proper references of data sources are mandatory. It includes the publication and new (meta)data shared alongside it. References of data citation should include the minimum identifier information, such as authors, dataset name, publication year, repository/publisher, and DOI. It guides readers to easily access the dataset.

 

Data repository

For the science continuity, authors are encouraged to preserve datasets permanently on a dataset repository. For students or researchers in universities, they could submit the (meta)data to their institutes’ repositories provided by the sponsor or the university. Scholars in special disciplines prefer to choose a discipline-specific data repository. Here are some examples.

 

If there is no mandatory repository in your discipline, scholars could submit datasets to a generalized dataset repository. Below are some examples of generalized repositories.