Disseminating your work

Last reviewed: October 2026. Official sources: ABCD docs: Acknowledgement, Responsible use, NBDC FAQs, and the DUC.

This page covers what you must do (and what you should do) when sharing ABCD findings in papers, preprints, posters, talks, and code.

Required: the short checklist

• Active DUC through submission, revision, and publication (DUC §5).

• Cite the release DOI in every paper, abstract (as space allows), poster, and talk (DUC §17).

• Include the ABCD acknowledgment statement (below).

• No cells smaller than 10 in any table, figure, or text (DUC §6). This includes demographic tables, crosstabs, and subgroup plots.

• No individual-level data, raw or derived, in supplements, repositories, or figures that could identify participants (DUC §4). If unsure, email nbdc@mail.nih.gov before publishing.

• A-DUC if you report American Indian/Alaska Native results separately (see Getting access).

• Register the publication with NBDC once it has a DOI or PMID (DUC §16), and list it in your annual DUC Progress Report.

Citing the data

Use the DOI for the specific release you analyzed (NBDC ABCD page):

Release	DOI
7.0	https://doi.org/10.82525/8f3w-5260
6.1	https://doi.org/10.82525/ab4q-qr87
6.0	https://doi.org/10.82525/jy7n-g441
5.1.1	https://doi.org/10.82525/wjkw-bp63

Note

The NBDC 7.0 page displays the DOI as “10.1.66.93/8f3w-5260”, which appears to be a typo. Its link resolves to 10.82525/8f3w-5260, which is the one to use.

The ABCD documentation is also versioned on Zenodo. Citing the docs version you used (see the citation page) helps readers find the exact variable definitions.

The acknowledgment statement

Copy the current, full text from the ABCD acknowledgment page rather than from an old paper. The award list and wording change. It includes:

1. “Data used in the preparation of this article were obtained from the Adolescent Brain Cognitive Development™ (ABCD) Study (https://abcdstudy.org), held in the NIH Brain Development Cohorts Data Sharing Platform…”

2. The list of NIH award numbers and a link to the federal partners.

3. A disclaimer that consortium investigators did not necessarily participate in your analysis, and that your views don’t necessarily reflect those of NIH or the ABCD consortium.

4. Conditional additions: a sentence about versioned releases plus the DOI, and extra statements if you use linked environmental data (NIEHS) or the Social Development substudy (NIJ).

Also follow the ABCD trademark guidance (e.g., “Adolescent Brain Cognitive Development℠ Study” / “ABCD Study®”).

Responsible reporting

From the ABCD responsible-use guidance:

• Don’t treat race/ethnicity as a proxy for environmental exposures such as racism, poverty, or neighborhood. Measure those directly where possible. See Cardenas-Iniguez & Gonzalez (2024).

• Be cautious with SES measures, and don’t rely on a single indicator.

• Use person-first, non-stigmatizing language (see NIDA’s Words Matter).

• Check DEAP’s per-variable data-consideration and responsible-use flags for every variable you report.

• Report effect sizes with context. Most ABCD effects are small, and that’s expected (Owens et al. 2021; Dick et al. 2021).

Recommended: open-science practices

Preregistration

The DUC (§15) strongly encourages preregistering planned and exploratory analyses (e.g., on OSF) before analysis. Because ABCD data are already collected, be explicit about what you have and haven’t looked at.

• OSF Preregistration; for secondary data, use the Secondary Data Preregistration template

• Consider a Registered Report. Several journals (e.g., Developmental Cognitive Neuroscience) accept them.

• Blinded / view-only preregistration links for peer review (search “view-only link” in OSF Help)

• Example: search OSF Registries for “ABCD”

Sharing code (strongly encouraged, DUC §16)

The DUC literally points to this community’s GitHub (github.com/now-i-know-my-abcd) as an example of code sharing.

• Put analysis code on GitHub, and archive a release on Zenodo or OSF to get a citable DOI.

• Never commit data. Add your data directory to .gitignore, and double-check that outputs (figures, .RData, notebooks with printed data frames) don’t contain individual-level data or small cells.

• Share your variable list and inclusion/exclusion steps. Variable names and metadata are not participant data.

• Record package versions (renv / sessionInfo() in R; requirements.txt / environment.yml in Python).

What to report in the methods

Lopez et al. (2024) audited 549 ABCD papers and found that only ~30% shared scripts and ~38% reported per-variable missingness. You can do better by reporting:

• Release number, NBDC DOI, and docs version

• Exact variable names (new 6.0+ names; old names too if helpful)

• How you derived your analytic sample (flowchart with Ns at each step, all ≥ 10)

• Missingness per variable and how it was handled

• How you handled family, site/scanner, and repeated-measures dependence

• Whether and how you used population weights

• A link to your code and preregistration

The ABCD responsible-use page links a reproducibility checklist that is handy before submission.

Posters, talks, and teaching

• The same rules apply: cite the DOI, include the acknowledgment (abbreviated if space requires, with a link), and keep all cells ≥ 10.

• You may show individual-level data in educational settings (e.g., a workshop) to people without a DUC, but you may not give them the data (DUC §5).

• Preprints are public disclosures, so treat them like publications.