Data Availability Policy

The requirement

Every research article published in BORJSS must include a data availability statement. It appears before the references and is published with the article. A manuscript submitted without one is returned for it at desk check.

The journal encourages authors to deposit their data openly, and requires that they say accurately where the data are and on what terms they can be obtained. Those are different obligations, and the distinction matters: social science data frequently cannot be shared, and a policy that demanded it would either exclude legitimate research or produce statements that are not true.

“Data are available on request” is accepted only where the authors mean it — see retention and requests below.

What counts as data

Data here means the material a reader would need to verify the findings — not only a spreadsheet of numbers.

• Survey responses, in raw and cleaned form;
• Interview and focus group transcripts, and field notes;
• Coding frames, codebooks and coded outputs;
• Questionnaires, instruments and interview schedules;
• Observational records and administrative extracts;
• Analysis scripts, syntax files and model specifications;
• Any variable definitions and transformations applied between raw collection and reported results.

Instruments deserve particular mention: publishing the questionnaire or interview guide is usually possible even where the responses are not, and it is often the single most useful thing an author can share.

Acceptable statements

Use whichever of these fits, adapted to the specifics. Where more than one applies — an open instrument with restricted responses — say so.

A statement that says nothing checkable — “data are available from the authors” with no contact, or “data are confidential” with no reason — is returned for revision.

When data may be withheld

These are legitimate reasons not to share, and the journal expects to see them regularly in social science research:

• Participants did not consent to sharing. Consent obtained for one study does not authorise redistribution. Sharing beyond those terms would breach the research ethics policy.
• Anonymisation is not achievable. In a small community, a named institution or a specialised professional group, removing names does not prevent identification.
• Sharing would put participants at risk — research on contested political or religious questions, on illegal activity, or in conflict-affected settings.
• The data belong to someone else — a government department, a company, or a repository whose licence forbids redistribution.
• Legal or contractual restriction, including data protection obligations.

Commercial sensitivity and an intention to publish further papers from the dataset are weaker reasons. The journal accepts a stated embargo period for the latter, but not indefinite withholding.

Qualitative and sensitive data

Qualitative material carries identifying detail in a way survey data usually does not: a transcript reveals a person through their circumstances, their turns of phrase and the events they describe, long after their name has been removed.

Full transcripts should therefore not be deposited openly unless participants specifically consented to that and the material genuinely cannot identify them. Better options, in rough order of preference:

• Deposit the instruments — interview guide, topic list, coding frame — which are almost always shareable;
• Deposit an anonymised subset or a set of redacted excerpts sufficient to show how codes were derived;
• Deposit under controlled access, where a repository releases material to named researchers under an agreement;
• Where none is possible, explain precisely why in the statement.

Consent forms should ask about future sharing at the point of collection. Asking then makes sharing possible later; not asking closes the option permanently.

Where to deposit

Use a recognised repository that issues a persistent identifier rather than a personal website, a cloud drive or a departmental page — those disappear. A suitable repository provides a DOI or accession number, long-term preservation, and a stated licence.

Discipline-specific and institutional repositories are preferred where available; general repositories such as Zenodo or the Dataverse network are acceptable, and both support controlled access where open deposit is not appropriate.

Small supporting files — a questionnaire, an appendix table — may instead be published as supplementary material with the article, where they are covered by the article’s CC BY licence.

Citing and licensing data

A deposited dataset is a research output and should be cited as one, in the reference list, with its authors, title, repository, year and DOI — both in your own article and by anyone reusing it. Citing data is how the effort of producing it becomes visible.

For the data themselves the journal recommends CC0, because attribution requirements stack awkwardly when datasets are combined; the norm of citation does the work that a licence condition otherwise would. CC BY is also acceptable. State whichever applies in the availability statement.

Code and analysis scripts

Where analysis was performed in software that produces a script or syntax file, depositing it is strongly encouraged even when the data cannot be shared. Code often reveals more about how results were produced than a prose method section can, and it is rarely confidential.

Deposit code in a repository that issues a DOI, state the software and version, and license it under a recognised open source licence. Where AI tools were used to generate analysis code, that is disclosed under the AI-assisted writing policy.

Retention and requests

Authors must retain the data underlying a published article for at least five years after publication, as the research integrity policy requires — including where the statement says the data are not publicly available.

Where the statement says data are available on request, the journal expects the authors to respond to reasonable requests from other researchers within a reasonable period, and to apply the criteria they stated. An unanswered request is a failure to meet a published commitment, and readers may report it to the editorial office at ceoborjss@gmail.com.

Editors may request the underlying data at any point during review, or after publication where a concern is raised. Refusal to supply data to the editor in the course of an investigation is itself treated as a serious matter under the publication ethics policy.

What the journal does

• At desk check — a manuscript without a statement is returned for one.
• At review — reviewers are asked whether the statement is consistent with the method described, and whether anything claimed to be unshareable plausibly is.
• At publication — the statement is published with the article, so the commitment is on the public record.
• After publication — where a statement turns out to be inaccurate, the journal publishes a correction amending it under the retraction and correction policy.

The journal does not decline manuscripts because the data cannot be shared. It declines to publish statements that are not true.