The review process is intended to ensure that each submission is technically sound, the dataset is described in a way that facilitates reuse, and that the manuscript meets the journal’s formatting and quality standards. Unlike traditional research articles, the focus is on data integrity, methodological clarity, and completeness of the dataset description rather than on hypothesis testing or groundbreaking scientific findings.

Reviewers should assess the dataset’s integrity, technical rigor, and the clarity of the accompanying description. Essential criteria include robust data generation methods, sufficient technical validation, completeness of data files and metadata, and suitability of the chosen repository. Reviews should emphasize improvements that enhance the reproducibility and reusability of the data.

IEEE Data Descriptions accepts three article types:

• Descriptor Articles: Detailed accounts that describe a single dataset including its collection, validation, and storage details.
• Collections Articles: Focus on groups of datasets, often from a shared event or initiative.
• Meta Articles: Articles that discuss data-related topics or comparisons among datasets rather than describing new datasets.

Each type has specific formatting instructions and section requirements (e.g., Abstract, Background, Collection Methods, etc.), which reviewers should use as a checklist when assessing completeness and adherence to guidelines.

Check that the manuscript follows the prescribed structure (e.g., Abstract, Background, Methodology, Validation, etc.) as outlined in the Author Guide. The description should be logically organized so that:

• Background: Provides context and explains the dataset’s value.
• Collection Methods and Design: Details how the data were acquired.
• Subsequent sections (e.g., Validation, Records and Storage) are comprehensive enough to allow replication and reuse.

Reviewers often note ambiguities in the narrative, insufficient detail in describing data collection or processing, and unclear organization. Constructive feedback should suggest rephrasing and restructuring for better flow and completeness.

Reviewers need to ensure that:

• Data are generated using robust, well-documented methods.
• The manuscript clearly describes all data-processing steps and quality control measures.
• Any technical validations (e.g., sensor accuracy, error rates) are convincingly supported with figures or tables.

If data-processing details or quality validations are insufficient, reviewers should request additional details or analyses. The focus should be on ensuring that other researchers can replicate the dataset’s generation and reuse the data effectively.

This question pertains to the internal comprehensiveness of the dataset. It asks whether all relevant aspects or elements within the scope of the dataset are fully represented. For example, in a dataset on written language recognition, adequate coverage means that all letters or syllables of that particular language are included. It does not imply that the dataset must cover every known language or broader external topics; rather, it should provide a complete and representative sample of the subject matter it intends to capture.

Examine whether figures, tables, and charts are clear and support the textual description. Verify that:

• All data files are described (including file formats and structures).
• Figures and tables are well-labeled, and statistical measures (if applicable) are appropriately reported.
• Supplementary materials are used to provide additional, non-redundant information.

Common suggestions include reformatting tables for clarity, improving figure resolution, providing detailed legends, and ensuring that any data summaries are directly tied to the manuscript’s claims.

For IEEE Data Descriptions, the focus is on the utility and quality of the dataset rather than its novelty in the traditional sense. A manuscript should clearly explain how the dataset can support reproducible science, enable further research, or provide enhanced metadata for data discovery.

Even if the dataset seems straightforward, descriptor articles are valuable when they reveal critical methodological details or provide comprehensive metadata that facilitates reuse. In cases of previously published data, the manuscript should offer new insights or significant updates that justify a new publication.

Ensure that the manuscript includes:

• Statements of ethical approval (if applicable)
• Documentation of informed consent and conflict of interest disclosures

Reviewers should verify that ethical standards—including data sharing policies and copyright/licensing terms—are clearly addressed in line with IEEE policies.

The process involves:

1. Initial Assessment: Editors check if the submission is within scope and meets quality standards.
2. Reviewer Selection: At least two independent reviewers are assigned.
3. Detailed Evaluation: Reviewers assess technical rigor, data quality, completeness of descriptions, and adherence to guidelines.
4. Editorial Recommendation: Based on reviewer feedback and the manuscript’s overall quality, a recommendation is made (Accept, Accept with minor revisions, Major revisions, or Reject).

Reviewers must immediately disclose any potential conflicts (e.g., past collaborations, competing interests) to maintain the integrity of the review process. This ensures an unbiased evaluation.

When authors submit a rebuttal, reviewers (and editors) should provide clear, constructive explanations to support the editorial process. It is advised that new editors consult with the Editor-in-Chief if uncertainty arises during this stage.

A good review should:

• Clearly identify strengths and weaknesses.
• Offer actionable suggestions for improvement.
• Provide detailed comments that reference specific sections of the manuscript.
• Focus on data quality, methodological transparency, and completeness of the dataset description rather than solely on the manuscript’s literary style.
• Follow the recommended review format (Accept, Accept with minor revisions, Major revisions, or Reject) as outlined in the Reviewer Guide.

Reviews should be thorough enough to guide the authors in enhancing their manuscript while also aiding the editor in making an informed final decision. Specificity regarding sections or figures and clear suggestions for additional details or analysis are encouraged.

All reviewers should familiarize themselves with the IEEE Data Descriptions Author and Editor Guides. These documents provide context on the journal’s scope, article formatting, and overall expectations. This ensures that your review is aligned with the journal’s mission to promote high-quality, reusable datasets and transparent data reporting.