Skip to main content
v2026.11,986 entries · CC-BY 4.0

Documenting and Citing Government Datasets in Academic Research

A complete guide to documenting and citing government datasets in academic manuscripts, methods sections, and open-science data availability statements.

Written and maintained by CASRAI Editorial Board

Published

On this page

Documenting a government dataset in an academic manuscript requires researchers to identify the originating agency as corporate author, record the release vintage or retrieval date, provide a persistent identifier or catalog link, and detail extraction parameters in both the methods section and the data availability statement. Unlike journal articles, public government datasets often lack digital object identifiers (DOIs) and undergo periodic administrative updates, requiring authors to maintain clear documentation across the manuscript text, supplementary query scripts, and bibliographic references to ensure full scientific reproducibility.

Ask CASRAI · free to tryQuestion about this guide? Ask CASRAI answers with linked sources. 2 free questions a day, no account.

Ask your first question freeSubscribe

Purpose of the section

Documenting government datasets within an academic paper establishes data provenance, enables independent methodological replication, and provides formal academic credit to public statistical bodies. National and municipal agencies collect vast repositories of empirical data, including decennial censuses, macroeconomic indicators, satellite telemetry, and disease surveillance. When researchers rely on these secondary data sources to test hypotheses or calibrate models, peer reviewers and readers require unambiguous traceability from published tables back to the exact version of the primary public record.

In contemporary research practice, citing public sector information is a core compliance requirement enforced by research funders and academic publishers. Frameworks like the White House Office of Science and Technology Policy (OSTP) public access guidance, the National Institutes of Health (NIH) Data Management and Sharing Policy, and the European Union Horizon Europe framework mandate transparency regarding secondary data inputs. While works created by officers of the United States federal government enter the public domain under 17 U.S.C. 105 and are exempt from domestic copyright, academic integrity standards distinguish between copyright permissions and scholarly citation obligations. Researchers reusing public domain datasets remain ethically obligated to credit originating agencies and document data origins to avoid plagiarism and maintain evidentiary integrity.

A central challenge in working with open public records is the administrative structure of government publishing. A single agency may produce dozens of discrete statistical releases, including raw microdata files, geographic shapefiles, and curated analytical reports. Authors often mistakenly cite an agency summary report when their analysis drew from an underlying microdata file. Formal documentation in the manuscript clarifies whether findings derive from primary agency measurements, summary tabulations, or derived variables produced through custom queries. Following established data citation standards ensures that every transformation applied to public data remains transparent and verifiable.

Structure (ordered list)

Comprehensive documentation of government data spans multiple distinct sections of a manuscript. To satisfy publisher data policies and peer review standards, researchers must incorporate public dataset details across four synchronized components:

  1. Methods Section (Data Sources and Variable Construction): Explicitly name the originating agency, parent department, dataset title, survey cycle or release vintage, geographic resolution, sampling design, and extraction timeframe. If the dataset was retrieved via an Application Programming Interface (API), record the base endpoint, query parameters, API version, and retrieval date. When variables undergo harmonization, weighting, or inflation adjustments, specify the exact formulas and reference indices used.
  2. Data Availability Statement (DAS): A standalone data availability statement is mandatory across peer-reviewed journals. This statement must declare that the dataset is publicly accessible, identify the hosting portal (such as Data.gov, Eurostat, or the UK Data Service), provide the permanent repository link or digital object identifier (DOI), disclose any credentialing requirements (such as Census Bureau restricted-use agreements), and state the governing licensing terms under which the data are redistributed.
  3. Bibliographic Reference List: Every government dataset utilized in the study must receive a full formal citation entry in the bibliography, formatted under the target journal’s citation style (such as APA 7th edition, Chicago Manual of Style 18th edition, or MLA 9th edition). Modern bibliographic standards dictate that datasets receive independent citation entries rather than informal web links in footnotes.
  4. Supplementary Code and Materials Archive: Computational reproducibility requires depositing data preparation scripts into an accessible institutional repository or open code archive. This includes automated download scripts (in R, Python, Stata, or SAS), API query parameters (with private keys redacted), variable transformation routines, and tabular data dictionaries documenting original agency column headers.

A short annotated example

To demonstrate how these structural elements operate in a peer-reviewed manuscript, consider three standard scenarios representing demographic aggregates, environmental physical observations, and international macroeconomic time-series.

Scenario 1: U.S. Census Bureau Demographic Data (APA 7th Edition)

In this scenario, a study investigates regional socioeconomic trends using the American Community Survey (ACS) 5-year estimates. The author documents the dataset in the Methods section, includes a Data Availability Statement, and provides a full bibliographic reference.

Methods Section Excerpt: Demographic characteristics and median household income figures were obtained from the U.S. Census Bureau American Community Survey 5-Year Estimates for the 2019-2023 release vintage (Table B19013, Median Household Income in the Past 12 Months). Data were extracted at the county level via the Census Data API on October 14, 2026, using the tidycensus package in R. All nominal dollar amounts were adjusted to constant 2023 dollars using the Bureau of Labor Statistics Consumer Price Index for All Urban Consumers (CPI-U).

Data Availability Statement Excerpt: Publicly available secondary datasets were analyzed in this study. The demographic and socioeconomic indicators were retrieved from the U.S. Census Bureau American Community Survey 2019-2023 5-Year Estimates. These data are public domain government records available without restriction via the U.S. Census Bureau data platform (https://data.census.gov) and through the Census Data API. The R script utilized to execute the automated data query and assemble the analytic sample has been deposited in an open repository at https://doi.org/10.0000/example.census.script.

Reference List Citation (APA 7th Edition):

U.S. Census Bureau. (2024). American Community Survey 2019–2023 5-year estimates (Table B19013) [Data set]. U.S. Department of Commerce. https://data.census.gov

Annotation of Citation Elements:

  • Author (U.S. Census Bureau): The specific operating agency responsible for data collection serves as the corporate author. In APA 7th edition, identify the sub-agency rather than solely the broad Cabinet department.
  • Publication Year (2024): The calendar year when the specific 5-year estimate release was published, not the year the survey began.
  • Title (American Community Survey 2019–2023 5-year estimates): Italicized title in sentence case, explicitly stating the survey duration and reference cycle.
  • Catalog/Table ID in Parentheses ((Table B19013)): Identifies the exact statistical matrix extracted from the repository.
  • Bracketed Content Type ([Data set]): Essential APA metadata element designating the resource type to differentiate it from textual reports or articles.
  • Publisher (U.S. Department of Commerce): The parent governing department. In APA 7th, if the group author is identical to the publisher, omit the publisher; when citing through a broader department or external catalog, provide the parent entity.
  • Location (https://data.census.gov): Direct, stable URL to the official platform. For more details on census specifics, see our dedicated guide on citing census data.

Scenario 2: Environmental Sensor Data from NOAA (Chicago 18th Edition)

When physical science or geospatial studies draw from atmospheric or oceanic observations, persistent identifiers and geospatial bounding parameters are vital for scientific integrity.

Methods Section Excerpt: Daily sea surface temperature anomalies were acquired from the National Oceanic and Atmospheric Administration (NOAA) National Centers for Environmental Information (NCEI). We examined the Daily Optimum Interpolation Sea Surface Temperature (DOISST) high-resolution dataset (Version 2.1) covering coordinates 20°N to 35°N and 65°W to 80°W for the observational window spanning January 1, 2020, through December 31, 2025.

Bibliography Entry (Chicago 18th Notes and Bibliography Style):

National Oceanic and Atmospheric Administration. 2024. Daily Optimum Interpolation Sea Surface Temperature (DOISST), Version 2.1. NOAA National Centers for Environmental Information. Accessed November 14, 2026. https://doi.org/10.5065/example.doisst.

In-Text Citation: (National Oceanic and Atmospheric Administration 2024)

Scenario 3: International Macroeconomic Data from the UK ONS (MLA 9th Edition)

Studies evaluating international comparative policy frequently utilize public sector data governed by national open government licenses.

Methods and Attribution Excerpt: Monthly consumer price inflation rates were drawn from the UK Office for National Statistics (ONS) Consumer Price Indices series (MM23). The data are reproduced under the terms of the UK Open Government Licence v3.0, which grants unrestricted worldwide reuse subject to formal acknowledgment: “Contains public sector information licensed under the Open Government Licence v3.0.”

Works Cited Entry (MLA 9th Edition):

United Kingdom, Office for National Statistics. Consumer Price Inflation, UK: Time Series (MM23). Office for National Statistics, 18 Oct. 2024, www.ons.gov.uk/economy/inflationandpriceindices/datasets/consumerpriceinflation.

Common mistakes

Reviewers and journal editors routinely encounter significant deficiencies in manuscripts relying on public data. Avoiding these six frequent errors will accelerate peer review and ensure compliance with archival standards:

  • Citing Top-Level Portal Homepages: Authors frequently write generic citations such as “Data retrieved from Data.gov” or provide the home URL “https://www.bls.gov”. Portals like Data.gov serve as catalog aggregators indexing hundreds of thousands of individual data distributions. Citing the homepage is equivalent to citing “the library” rather than a specific book. Citations must point directly to the individual dataset record, permanent repository landing page, or documented DOI.
  • Conflating Analytical Reports with Underlying Datasets: Statistical agencies release press bulletins, research working papers, and annual summaries describing their data collections. Citing an analytical report when your study extracted raw microdata confuses readers regarding whether you conducted independent statistical modeling or merely quoted secondary summary statistics. Always cite both: cite the methodology report if discussing survey mechanics, and cite the dataset itself as an independent empirical object.
  • Omitting Release Vintages and Revision Dates: Government agencies frequently revise historical series to incorporate updated seasonal adjustments, benchmark data, or modified population controls. Macroeconomic indicators from the Bureau of Economic Analysis (BEA) are published as advance, second, and third estimates, and undergo annual comprehensive revisions. If an author fails to record the exact release vintage or date of access, subsequent researchers attempting replication will generate differing numerical outputs.
  • Assuming Public Domain Negates Citation Obligations: While U.S. federal data are not subject to copyright protection under 17 U.S.C. 105, copyright law and academic citation ethics govern distinct domains. Reusing open public records without explicit scholarly attribution constitutes academic misconduct. The fundamental data citation principles established by international scientific bodies stipulate that all empirical inputs must receive formal credit regardless of their legal copyright status.
  • Failing to Disclose Custom Filtering, Weighting, and Subsetting: Government surveys typically employ complex multistage probability sampling. When researchers extract subsets—such as restricting an American Community Survey sample to full-time employed respondents between ages 25 and 54—the specific filtering logic, replicate weight adjustments, and strata definitions must be fully documented in the Methods section. Without this metadata, independent replication is impossible.
  • Mismatched Agency Authorship Hierarchies: Academic style guides maintain specific rules for corporate and group authors. A common defect is listing the parent Cabinet department (such as the U.S. Department of Health and Human Services) while omitting the specific operating division (such as the Centers for Disease Control and Prevention) that collected and curated the data. Citing at the sub-agency level ensures proper institutional provenance and improves search discovery.

Checklist

Before submitting a manuscript that utilizes government datasets to a journal or preprint server, verify each item on this editorial checklist:

  1. Originating Agency Verified: The formal corporate author is identified at the specific bureau or agency level (e.g., U.S. Bureau of Labor Statistics, Eurostat, UK Office for National Statistics) rather than solely an overarching parent department or hosting web portal.
  2. Dataset Title and Series ID Recorded: The title reflects the exact formal name of the data product, including official survey acronyms, table numbers (e.g., Table B01001), series keys, or catalog identifiers.
  3. Vintage and Temporal Bounds Specified: The survey year, baseline census wave, longitudinal vintage, or specific time-series range is explicitly recorded in the Methods text and reference entry.
  4. Persistent Link or DOI Included: The citation provides a permanent digital object identifier (DOI), a direct catalog URL, or an established archival handle rather than a broad institutional homepage.
  5. Retrieval Date Documented: If the data source consists of an active, continuously updated database or an interactive query interface lacking static versioned files, the access date is recorded in both the bibliography and the Data Availability Statement.
  6. Licensing Terms and Attributions Stated: The manuscript confirms the reuse terms—whether U.S. public domain under 17 U.S.C. 105, UK Open Government Licence v3.0, Creative Commons, or a specialized data use agreement—and includes any mandatory statutory attribution phrases.
  7. Data Availability Statement Completed: The standalone Data Availability Statement clearly discloses the availability status of all public datasets, specifies public repositories, and directs readers to persistent access links.
  8. Query Parameters and Analysis Scripts Archived: Any custom computer code used to query public APIs, merge administrative files, apply survey weights, or recode variables has been deposited into an accessible research repository or formatted as supplementary files.

Discipline differences

Methodological expectations and reporting standards for government datasets vary significantly across academic fields. Researchers must conform to the specialized evidentiary conventions of their respective disciplines:

Economics, Finance, and Labor Studies: Economists rely heavily on administrative records and time-series data from agencies such as the Bureau of Labor Statistics (BLS), Bureau of Economic Analysis (BEA), the Federal Reserve (FRED), and Eurostat. In top economics journals, replication packages are mandatory. Authors must provide the exact series identifiers (e.g., BLS Series ID CUUR0000SA0 for headline CPI) and deposit complete Stata, R, or Python do-files that execute the transformation from raw agency flat files to final regression tables. Because economic series are subject to regular benchmark revisions, stating the exact release vintage is standard practice.

Public Health, Epidemiology, and Clinical Research: Health researchers frequently utilize public surveillance databases such as the CDC WONDER system, the National Health and Nutrition Examination Survey (NHANES), and Centers for Medicare and Medicaid Services (CMS) public use files. In this discipline, ethical data governance is paramount. Even when working with publicly released, de-identified datasets, authors must document compliance with institutional review board (IRB) exemptions and disclose whether the study used open public-use files or restricted-access microdata accessed through secure enclaves requiring specific Data Use Agreements.

Environmental Science, Meteorology, and Oceanography: Geoscientists working with data from NASA Earthdata, the U.S. Geological Survey (USGS), and the National Oceanic and Atmospheric Administration (NOAA) must provide rigorous spatial and temporal coordinate boundaries. Scientific reporting in these fields requires specifying spatial resolution (e.g., 0.25-degree grid cells), projection coordinate systems, instrument calibration versions, and platform identifiers (such as satellite orbits or terrestrial streamgage numbers). Compliance with standard geospatial metadata formats, such as ISO 19115 and DCAT-US geospatial schemas, is routine.

Political Science, Sociology, and Public Administration: Social scientists drawing on government election returns, federal campaign finance filings from the Federal Election Commission (FEC), and municipal crime data must pay close attention to boundary harmonizations and legal definitions. Political science journals emphasize documenting changes in municipal reporting thresholds, statutory reporting categories, and redistricting boundaries that alter spatial units across electoral cycles.

Regardless of discipline, the broader open-science movement emphasizes the adoption of FAIR data principles—ensuring research assets are findable, accessible, interoperable, and reusable—and the deposit of research assets in recognized data repository environments. The comparative data table below summarizes primary federal and international government data repositories, detailing their governing agencies, metadata architectures, access mechanisms, licensing terms, and persistent identifier conventions.

Frequently asked questions

Do I need permission to use government datasets in a published research paper?

In most instances, no formal permission is required. Works created by the United States federal government are placed in the public domain under 17 U.S.C. 105, meaning they can be freely downloaded, analyzed, visualized, and published without licensing fees or written authorization. Similarly, data published by the UK government under the Open Government Licence (OGL v3.0) or the European Union via open data directives are pre-cleared for academic and commercial reuse. However, you must always provide formal citation and attribution, and if you are using restricted-use government files that contain sensitive personal or commercial information, you must adhere to the data use agreements executed with the statistical agency.

How do I cite a government dataset if it does not have a DOI?

Many government datasets do not carry digital object identifiers (DOIs). When a DOI is unavailable, provide a direct, stable URL pointing to the dataset’s official catalog landing page or data repository entry. In your bibliographic citation, include the unique dataset identification number, table number, or series code in parentheses immediately following the title (e.g., Table B01001 or Series CUUR0000SA0), followed by the bracketed description [Data set], the publishing agency or platform, and the URL. If the data were obtained through an interactive database query that generates temporary session links, cite the main portal landing page and detail the query parameters in your Methods section.

What is the difference between citing a government report and citing a government dataset?

A government report is a textual document containing narrative analysis, policy evaluations, or pre-computed statistical summaries authored by agency staff or contractors. A government dataset is the underlying collection of raw or structured numeric observations, microdata records, or tabular matrices used to conduct empirical calculations. If your study performs independent statistical analysis, regressions, or visualizations on public numbers, you must cite the dataset itself. If you are quoting an agency’s official policy conclusion or descriptive text, cite the published report.

How should I document data retrieved through a government API?

When retrieving data programmatically via an agency API, document the protocol in both your Methods section and your supplementary code archive. State the base API endpoint, the API version, the date and time of the query execution, and the exact query parameters or endpoints queried (such as specific geography codes, survey years, and variable IDs). In your bibliography, cite the underlying dataset, noting the API access method. Never publish your personal private API authentication keys or security tokens in the manuscript or public code repositories.

Where does the dataset citation go in an academic manuscript?

Dataset citations must appear in two distinct locations: within the main text (in the Methods section and the Data Availability Statement) and in the final Reference list or Works Cited section. In-text mentions should use standard parenthetical or narrative citation formatting (e.g., U.S. Census Bureau, 2024), while the final bibliography provides the complete bibliographic entry with author, title, version, catalog identifier, [Data set] label, and persistent URL or DOI. This dual placement ensures that indexing databases can track citations to government data assets.

Sources

  • Project Open Data. (2026). DCAT-US Schema and Metadata Resources. General Services Administration and Federal CDO Council. Checked 2026-11-14. https://resources.data.gov/resources/dcat-us/
  • Data.gov. (2026). Data Policy, Governance and Catalog Services. U.S. General Services Administration. Checked 2026-11-14. https://data.gov
  • U.S. Census Bureau. (2026). Data Access, API Documentation, and Archival Data Tools. U.S. Department of Commerce. Checked 2026-11-14. https://www.census.gov/data.html
  • FORCE11. (2014). Joint Declaration of Data Citation Principles. FORCE11 Data Citation Synthesis Group. Checked 2026-11-14. https://force11.org/info/joint-declaration-of-data-citation-principles-final/
  • American Psychological Association. (2020). Publication Manual of the American Psychological Association (7th ed.). Reference Examples: Data Sets, Software, and Tests (Section 10.9). Checked 2026-11-14. https://apastyle.apa.org/instructional-aids/reference-examples.pdf
  • The National Archives. (2026). Open Government Licence for Public Sector Information (OGL v3.0). Her Majesty’s Stationery Office / The National Archives. Checked 2026-11-14. https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/
  • Eurostat. (2026). Data Navigation Tree, Dissemination Database, and SDMX Metadata. European Commission. Checked 2026-11-14. https://ec.europa.eu/eurostat/web/main/data/database

Data: Major Government Data Repositories for Academic Research: Metadata and Access Matrix

Comparative synthesis of primary open government data platforms across the United States, European Union, and United Kingdom, detailing their administrative authority, underlying metadata standard, programmatic API accessibility, licensing terms, and standard citation models as documented in official repository specifications.

Period covered: 2024/2026. Coverage: United States, United Kingdom, European Union.

Major Government Data Repositories for Academic Research: Metadata and Access Matrix
Repository or PortalGoverning Agency / JurisdictionMetadata StandardAccess MethodLicensing ModelPersistent Identifier Type
Data.govU.S. General Services Administration (Federal)DCAT-US Schema (Project Open Data v1.1 / v3.0)Bulk download, REST API (CKAN API)U.S. Public Domain (17 U.S.C. 105)Catalog harvest ID / agency record ID
data.census.govU.S. Census Bureau (Department of Commerce)Census Data API Schema & DCAT-USInteractive query tables, bulk CSV, REST APIU.S. Public Domain (17 U.S.C. 105)Table ID (e.g. B01001) / Survey vintage
BLS Data PortalU.S. Bureau of Labor Statistics (Department of Labor)BLS Public Data API v2.0 JSON & DCAT-USText flat files, interactive tools, REST API v2.0U.S. Public Domain (17 U.S.C. 105)Series ID (e.g. CUUR0000SA0)
NOAA NCEI Data ArchivesNational Oceanic and Atmospheric AdministrationISO 19115-2 / FGDC-STD-001-1998 & DCAT-USHTTP/FTP bulk download, Climate Data Online APIU.S. Public Domain (17 U.S.C. 105)Digital Object Identifier (DOI)
data.gov.ukUK Cabinet Office / National ArchivesCKAN / DCAT-AP profileBulk download, CKAN APIOpen Government Licence v3.0 (OGL)Catalog URI / Department dataset ID
Eurostat DatabaseEuropean Commission (Statistical Office of the European Union)SDMX (Statistical Data and Metadata eXchange)Interactive data browser, bulk download TSV, SDMX 2.1 APIOpen reuse with mandatory attribution (Creative Commons Attribution 4.0)Eurostat Dataset Code (e.g. prc_hicp_midx)

Sources

Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about Documenting and Citing Government Datasets in Academic Research

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →