Open Data and Community Data: How to Share Participatory Research Results Responsibly

Citizen science & data · By the Common Inquiry editorial team · Published 2026-10-26

Two young people outdoors, one shading her eyes as she looks up, during a water monitoring field day — Students taking part in a volunteer water monitoring programme in Iowa. US government photo (USDA Natural Resources Conservation Service), public domain.

Participatory research produces data that belong, in a moral sense at least, to the people who provided them and often helped collect them. Sharing those data openly can multiply their value; sharing them carelessly can harm the very people they describe. This guide sets out the main choices a project faces, from giving results back to participants to publishing a dataset in a public repository.

Three audiences, three formats

A participatory project usually has to share data in at least three ways:

With participants and the wider community, in formats people can understand and use: meetings, posters, short reports, maps, translated summaries.
With decision-makers, as concise evidence linked to specific requests.
With other researchers and the public, as documented datasets that can be checked and reused.

Each audience needs a different level of detail, and each raises different risks.

Returning results to participants

"Report-back" is the practice of giving participants their own results and the study's overall findings. Environmental health researchers have studied it closely. A 2007 article by Julia Green Brody and colleagues in the American Journal of Public Health, "Is it safe? New ethics for reporting personal exposures to environmental chemicals", argued that participants generally have a right to know their own results, even when the health meaning is uncertain, and that report-back should be planned with community partners. Practical lessons from that literature include:

explain what a result means and does not mean, including uncertainty;
give comparison points (other participants, official values) where they exist;
say what people can do, if anything, to reduce a risk;
offer a way to ask questions, in person where possible.

Collective results can be shared through "data walks", in which residents move between posters of findings in small groups and discuss what they see. Co-interpretation sessions often catch errors and add explanations that researchers missed.

What "open" means

Open data are data that anyone can freely access, use, modify and share, subject at most to requirements to attribute the source and keep the same openness. In practice, publishing open data means choosing a licence. Common choices are Creative Commons Attribution (CC BY), the CC0 public domain dedication, and, for databases, the Open Database Licence (ODbL). Without a licence, others cannot be sure what they are allowed to do.

The UNESCO Recommendation on Open Science (2021) supports open data, but it also endorses the idea that data should be as open as possible and as closed as necessary, recognising legitimate reasons to restrict access, including privacy, security and the rights of Indigenous peoples. The FAIR principles (findable, accessible, interoperable, reusable) do not require data to be open either: data can be FAIR and still restricted.

Protecting people in published data

Community research often concerns small areas and small groups, which makes re-identification easier than in a national survey. Standard precautions include:

Removing direct identifiers such as names, addresses, phone numbers and photographs of faces.
Aggregating locations to blocks, streets or neighbourhoods rather than exact addresses or GPS points.
Suppressing small numbers: official statistics agencies routinely withhold or round counts below a threshold so that individuals cannot be singled out.
Checking combinations: age, occupation and street together may identify someone even if each alone does not.
Removing free-text answers or reviewing them line by line, since people often mention names and places.

Where data cannot be safely anonymised, controlled-access repositories let approved researchers use them under agreements. In the UK, the UK Data Service, funded by the Economic and Social Research Council, offers open, safeguarded and controlled access levels and guidance on preparing qualitative and quantitative data for archiving.

When data should not be open

Data about Indigenous peoples, unless the community has agreed; see our article on Indigenous data sovereignty.
Locations of vulnerable people, sensitive sites or rare species that could be targeted.
Information that could expose participants to retaliation, for example tenants' complaints identifiable by landlords.
Data that participants were promised would remain confidential. Consent forms should say clearly whether data will be shared, with whom and in what form.

Repositories and documentation

For data that can be shared, a recognised repository is better than a project website, which may disappear when funding ends. Options include discipline repositories (for example, biodiversity records through the Global Biodiversity Information Facility), national data archives, university repositories and general-purpose services that issue persistent identifiers (DOIs). Whatever the choice, publish documentation with the data: the protocol, a codebook explaining every variable, the quality checks applied (see data quality in citizen science), known limitations, and a clear licence and citation.

Write a data agreement early

Disputes about data are among the most common reasons partnerships break down. A short agreement, signed before data collection, should cover:

Ownership and stewardship. Who owns the data, who looks after them, and where they are stored.
Access. Who can see raw data during and after the project.
Use and reuse. What the data may be used for, and whether new uses need approval.
Publication. Whether community partners review outputs before publication, and how disagreements are resolved.
Retention. How long data are kept, and whether they are archived, returned or destroyed.

These questions apply equally to large citizen science platforms, where volunteers should be able to find a clear statement of how their contributions are licensed and used.

Sources

UNESCO, Recommendation on Open Science, 2021.
Julia Green Brody et al., "Is it safe? New ethics for reporting personal exposures to environmental chemicals", American Journal of Public Health, vol. 97, no. 9, 2007.
Mark D. Wilkinson et al., "The FAIR Guiding Principles for scientific data management and stewardship", Scientific Data, vol. 3, 2016.
UK Data Service, guidance on data access levels and on anonymising quantitative and qualitative data.
Open Knowledge Foundation, The Open Definition.