Citing a dataset that supports a research paper follows the same core logic as citing any source: you need to give readers enough information to locate the exact version you used, and you should follow the citation style required by your target journal or institution. In practice, this means including the dataset’s author(s), title, version or date, the repository or platform where it’s hosted, and a persistent identifier like a DOI or accession number.
The Standard Elements of a Dataset Citation
Regardless of whether you use APA, MLA, Chicago, or IEEE, a complete dataset citation typically includes:
- Author(s) or creator(s) – the person or organization who compiled the data
- Title of the dataset – as it appears in the repository
- Version or edition – if the dataset has been updated, always cite the specific version you used
- Publisher or repository – such as Zenodo, Dryad, Figshare, ICPSR, or a university data archive
- Year of publication or release
- Persistent identifier – a DOI, handle, or accession number (e.g., from NCBI or ArrayExpress)
Here is an example in APA 7 format:
Smith, J., & Lee, A. (2023). Global temperature anomalies, 1900–2022 (Version 2.1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.1234567
Where to Place the Citation
You have two complementary options:
- In-text citation – mention the dataset parenthetically where you first reference it, just like a journal article (e.g., “(Smith & Lee, 2023)”).
- Reference list entry – include the full citation in your bibliography, alphabetized with your other sources.
Additionally, many journals now require a data availability statement in the manuscript, separate from the reference list. This is a brief paragraph stating where the data can be accessed and under what conditions.
What If the Dataset Has No DOI?
Not every dataset is formally published. If you downloaded data from a website or a public repository without a persistent identifier, cite the URL and access date instead. If the data is part of a supplementary file attached to a paper, cite the paper itself and mention the supplementary material.
A Practical Tip
Using a citation manager or an academic assistant tool can save you from manual formatting errors. For example, WisPaper’s TrueCite automatically verifies your references and checks that your dataset citations follow the exact style your journal requires — which is especially helpful when you’re juggling dozens of sources across different formats.
Finally, always check the repository’s own “How to Cite” section — most major data archives generate a ready-made citation string for you. Copying that directly ensures you don’t miss version numbers or licensing details that could trip up peer reviewers.
