RESEARCH DOSSIER
Community Research Ethics, Knowledge Sovereignty, and Accountability
Abstract. Community research ethics investigates how scientific inquiry creates duties among individual participants, communities, universities, funders, data stewards, and users of research evidence. This critical narrative synthesis examines historical participatory research, evolving consent and authorization, governance of interpretation, Indigenous knowledge and data sovereignty, the First Nations principles of OCAP, the CARE and FAIR frameworks, privacy and re-identification, benefit sharing, and research accountability. The inquiry evaluates empirical and theoretical work on community-based participatory research, especially partnership mechanisms and measurement limitations, alongside official ethical and data-governance guidance. It distinguishes individual consent from collective responsibilities and Indigenous political authority, documents limits to causal generalization from observational partnership studies, and develops auditable research governance questions. It also examines digital communities and artificial systems as contemporary contexts for data reuse and delegated interpretation. The review draws on identified primary studies, interpretive reviews, and policy sources; it includes neither original participant data nor an exhaustive systematic-review or meta-analytic procedure.
Keywords: community research ethics; knowledge sovereignty; Indigenous data sovereignty; community-based participatory research; research accountability; participatory governance; OCAP; CARE principles; FAIR data; informed consent; collective authority; benefit sharing; community psychology; research power; ethics of artificial systems
Community Research Ethics as a Psychological Research Object
Community research ethics concerns the distribution of authority, obligation, knowledge, risks, and benefits when psychological inquiry involves people as members of social groups and institutions. Its central unit is the relationship among participants, collectivities, investigators, sponsors, and users of evidence. An individual interview may produce claims about an identifiable neighborhood, language community, or Indigenous nation; interpretation and disclosure can therefore affect people who never signed a consent form. Ethical inquiry must trace the life of information from question selection through recruitment, analysis, reuse, and public representation. The historical community-based research literature makes participation in these decisions an explicit feature of research practice Israel et al. (1998) Cargo & Mercer (2008) National Commission (1979).
The specific questions are: who may define a research problem, who can legitimately represent affected constituencies, what duties arise around collectively significant knowledge, how can agreement be revised, and who answers for unintended consequences? The answer requires distinguishing personal agency, community organization, institutional authority, and the special status of Indigenous self-determination. One community may contain contradictory interests, unequal access to decisions, overlapping membership, and people with no desire to participate. An ethics that recognizes communities must examine internal differences and dissent as seriously as relationships with external institutions TCPS 2 (2022) Carroll et al. (2020) Hudson et al. (2020).
This monograph develops a critical narrative synthesis of landmark participatory research, empirical partnership studies, research ethics scholarship, Indigenous data governance frameworks, and official guidance. Sources are cited for claims they can establish: policy documents demonstrate obligations within their jurisdictions, interviews and partnership studies describe observed practices, and reviews support cautious conclusions across heterogeneous settings. No new human-participant data were collected; this is neither a preregistered systematic review nor a meta-analysis. Its arguments distinguish descriptive evidence about research practice from normative conclusions about justified duties and recommended governance designs Jagosh et al. (2012) Oetzel et al. (2018) Guillemin & Gillam (2004).
The Community as a Heterogeneous and Changing Referent
A community may be organized by territory, kinship, language, shared experience, institution, political authority, or digital interaction. These bases rarely coincide. Survey boundaries can be administrative while lived affiliation is relational; research on a population segment may produce reputational consequences for a wider group. Defining the referent in advance is therefore essential for determining who is affected and who could reasonably be consulted. The governance problem changes when a study concerns one formally constituted nation, a diffuse health-identity network, or residents of an unstable neighborhood Israel et al. (1998) TCPS 2 (2022).
Representation is a testable institutional claim rather than a ceremonial label. Investigators can document how community partners were selected, whom they consult, how dissent is recorded, how frequently membership changes, and whether underrepresented residents can challenge decisions. Consultation by elected representatives, cultural knowledge holders, service providers, and informal networks may generate different forms of legitimacy. No single representative automatically has authority to consent on behalf of every individual; some knowledge involves particular custodians and specific collective obligations AIATSIS (2020) TCPS 2 (2022) FNIGC (n.d.).
Psychological analysis adds questions about perceived procedural justice, trust, power, belonging, and social identity. People may support a partnership's goals while rejecting its chosen spokesperson, or may disclose sensitive experience privately while refusing collective publication. Researchers should define which findings refer to persons, groups, institutions, or nested levels. These distinctions constrain interpretation and help avoid ecological fallacies, stereotyping, and the concealment of within-community variability. Community governance must be sufficiently plural to record disagreement without exposing dissenters to retaliation Buchanan et al. (2007) Guillemin & Gillam (2004).
Normative Foundations: Persons, Collectivities, Justice, and Reciprocity
The Belmont Report articulated respect for persons, beneficence, and justice within the United States' framework for protecting human research participants. In community psychology, applying these principles requires attention to collective spillovers and institutional conditions shaping apparently individual decisions. Respect includes comprehensible information and meaningful refusal; beneficence includes foreseeable risks of stigma or misuse of findings; justice includes who bears recruitment burdens and who receives the products of research. The extension to communities is a reasoned ethical interpretation of the familiar principles, not a statement that Belmont itself establishes an unrestricted collective veto National Commission (1979) HHS OHRP (n.d.).
Reciprocity adds an assessment of what partners contribute and receive. Local knowledge, introduction to trusted networks, interpretation, translation, and labor have value independent of academic authorship. A project that promises eventual benefit but budgets no time for community partners can reproduce distributive inequality under participatory language. Agreements about compensation, capacity development, reporting, and continuing access to findings make claims of reciprocity auditable. Participatory scholarship identifies mutual benefit and the treatment of power differences as recurring challenges Cargo & Mercer (2008) Wallerstein & Duran (2010) Israel et al. (2026).
Collective responsibilities vary by context. Indigenous peoples possess distinct histories, political and legal standing, and systems of knowledge stewardship; these rights should not be reduced to the general idea that any recruited social group owns every datum collected about it. The AIATSIS Code, First Nations OCAP principles, and Canadian TCPS 2 Chapter 9 articulate specific governance duties and interpretive requirements. Sound practice begins with the authority actually involved, the relevant jurisdiction, and the applicable community protocols AIATSIS (2020) FNIGC (n.d.) TCPS 2 (2022).
Historical Formation of Participatory Research Ethics
Community-engaged psychological research emerged from several lines of inquiry, including action-oriented social research, social ecology, health equity, and critiques of externally imposed expertise. Histories of colonial extraction and the political use of knowledge sharpened questions about who sets research priorities and who can control representation. Linda Tuhiwai Smith's Decolonizing Methodologies provided an influential analysis of how research practices participated in colonial relationships and how Indigenous research agendas can organize knowledge production on different institutional terms Smith (2021) Kukutai & Taylor (2016).
Israel, Schulz, Parker, and Becker's 1998 review described community-based research partnerships through participation across research stages and discussed conditions facilitating or obstructing their work. Cargo and Mercer subsequently organized challenges into an integrative participatory research practice framework. Wallerstein and Duran developed the argument that community-based participatory research could contribute to intervention science by incorporating community knowledge and addressing imbalances of power. These texts are foundational reviews and conceptual syntheses rather than a single experimental proof of general effectiveness Israel et al. (1998) Cargo & Mercer (2008) Wallerstein & Duran (2010).
An important later development is the formal articulation of governance rights over data themselves. The Indigenous data sovereignty scholarship edited by Kukutai and Taylor, the CARE principles, OCAP, and research ethics frameworks for Indigenous communities connect methodology with authority over collection, custody, interpretation, access, and reuse. Israel and colleagues' 2026 review documents expansion in partnerships, evaluation instruments, and attention to structural change. Historical continuity therefore lies in shifting the locus of evaluation from consent at entry toward accountability throughout the research relationship Kukutai & Taylor (2016) Carroll et al. (2020) Israel et al. (2026).
Research Method and the Hierarchy of Evidence
Evidence about community research ethics comes from different designs with different inferential reach. An official code can establish the content of an ethical standard; it cannot, by itself, prove that institutions comply. A partnership survey can test associations between governance practices and perceived collaboration quality, yet its respondents may share expectations or incentives that bias reports. Ethnographic and qualitative investigations can expose unanticipated obligations and explain mechanisms through which exclusion or trust develops, although analytic transfer depends on context and transparent interpretation. A realist review explores context–mechanism–outcome patterns without assuming a uniform treatment effect Jagosh et al. (2012) Guillemin & Gillam (2004) Oetzel et al. (2018).
Research on participatory partnerships faces selection and counterfactual problems. Communities and academics choosing intensive collaboration may already possess stronger networks, stable funding, or experiences that distinguish them from comparison sites. The cost of deliberation can vary by subject, time horizon, and available authority. For credible claims about outcomes, researchers should specify processes, comparison conditions, measurement invariance, time-ordering, missing data, and whether the study measured perceived partnership quality or actual health and policy change. Oetzel and colleagues' 2018 model test informs hypotheses about relationships among context, partnership processes, and outcomes; its observational associations require causal restraint Oetzel et al. (2018).
The present synthesis uses purposive landmark and contemporary source retrieval, comparison of published abstracts and accessible full texts, and examination of official research-governance statements. It distinguishes peer-reviewed empirical findings, interpretive reviews, ethical arguments, and jurisdiction-specific rules. Numerical pooled effects are excluded because a systematic search, dual screening, study-level extraction, and risk-of-bias assessment were not conducted. Claims about policy or general effectiveness remain proportional to the supporting evidence Israel et al. (2026) Jagosh et al. (2012).
Individual Consent in a Collective Research Relationship
Informed consent remains anchored in a particular person's understanding and voluntary choice, even when research is negotiated collectively. Participants need clear accounts of purposes, likely uses of recordings or tissue, foreseeable risks, withdrawal possibilities, compensation, confidentiality, and dissemination. Community consultation can improve language, recruitment design, and cultural relevance, while personal refusal retains independent moral significance. Researchers cannot infer the agreement of every member from a council's endorsement or the signature of a research partner National Commission (1979) TCPS 2 (2022).
Consent becomes more complex when data can be reused in unforeseen ways. A participant may understand an interview about local services but not anticipate combining transcripts with location, ancestry, network connections, and machine-learning features. Consent processes must explain realistic limits of confidentiality and the distinction between initial research, later secondary analysis, and public reuse. The agreement may also establish an independent route to ask questions or withdraw from further contact, while recognizing that deleting already published aggregate findings may be impossible TCPS 2 (2022) Carroll et al. (2020) Hudson et al. (2020).
The ethics of participation extends over time. When political conditions change, a community body dissolves, new representatives are elected, or a study discovers unexpectedly sensitive patterns, previous approval may require renewed consultation. Ongoing communication can reveal threats to interpretation that were invisible at recruitment. Guillemin and Gillam's account of ethically important moments clarifies why procedural review is necessary but insufficient for resolving the judgments that arise during ordinary fieldwork Guillemin & Gillam (2004).
Community Authorization and the Limits of Delegated Authority
Community authorization refers to approval or guidance from a body or set of people recognized as competent regarding a particular research relationship. Its legal and ethical form differs across contexts. TCPS 2 Chapter 9 requires researchers in specified Canadian Indigenous contexts to seek community engagement and to respect relevant customs and governance structures; research on lands under Indigenous jurisdiction may entail review by community and institutional bodies. The same chapter distinguishes forms of engagement and recognizes that communities can decline active participation TCPS 2 (2022).
An accountable authorization process documents whom a negotiating body represents, the categories of knowledge covered, how objections are heard, and which decisions remain reserved to individuals. Researchers must avoid transferring authority wholesale to whichever contact is easiest to recruit. Women, youth, people living outside a territory, persons with disabilities, and politically marginalized residents may have different stakes in disclosure. The ethical objective is a governance arrangement responsive to those differences, with explicit jurisdiction and transparent review of contested decisions AIATSIS (2020) FNIGC (n.d.).
Disagreement can become an important observation about the research setting, yet investigators should not exploit conflict to obtain favorable permission. Institutional ethics boards may need to consider whether pressure from employers, local authorities, or resource providers undermines individual choice. Conversely, a university cannot invoke private consent as a universal justification for disregarding established collective responsibilities around protected knowledge. Practical resolution requires negotiated scope and a record of decisions rather than presuming that either side always overrides the other Buchanan et al. (2007) TCPS 2 (2022).
Power, Funding, Labor, and Academic Extraction
Power asymmetry in community research can be observed through control over the budget, grant eligibility, employment contracts, data storage, research questions, analytic software, authorship decisions, and access to publication venues. A partner may be called a co-investigator while the university retains exclusive control of approvals and disbursements. Assessing participation therefore requires examining decision rights and resources, beyond meeting attendance and favorable survey responses. The participatory research literature repeatedly identifies power sharing, capacity, and institutional commitments as conditions that shape partnership processes Cargo & Mercer (2008) Israel et al. (2026).
Compensation can reduce participation costs but also create conflicts of interest. Community researchers often contribute local legitimacy, translation, emotional labor, safety advice, and interpretation that cannot be replaced by occasional consultation. Funding should specify fair pay, employment protections, travel and accessibility costs, and post-grant responsibilities. The distribution of overhead, indirect costs, equipment, and ownership of research products affects whether a partnership can continue once a project ends. These are governance decisions with consequences for who can accumulate scientific expertise Wallerstein & Duran (2010) Jagosh et al. (2012).
An extraction analysis asks whether locally generated knowledge travels to academic repositories without equivalent authority, credit, or return. The possibility of extraction does not establish that all outsider research is exploitative; it generates concrete questions about provenance, consent, benefit sharing, knowledge custodianship, and remedies. Indigenous scholarship makes the historical dimension of these questions visible, especially where research has been linked to dispossession or the denial of a people's knowledge institutions Smith (2021) Kukutai & Taylor (2016).
Trust, Procedural Justice, and Institutional Memory
Trust in research reflects expectations about competence, truthfulness, fair treatment, and institutional responsiveness, shaped by earlier experiences of participation. A signed agreement cannot erase memories of broken promises, unexplained sampling, or stigmatizing reports. Researchers can build justified confidence through predictable procedures, available records, stable points of contact, and credible avenues for complaint. Perceived fairness and experienced influence should be measured separately from satisfaction with a researcher's personality Israel et al. (1998) Oetzel et al. (2018).
Trust is frequently uneven within a community. Long-standing partner organizations may have close ties to universities while residents outside professional networks remain skeptical or uninformed. Repeated contact can produce genuine collaboration or dependency on a particular gatekeeper. Studies should therefore sample nonparticipants and dissenting voices where ethically feasible; reliance only on active partners risks selection bias and a circular conclusion that collaborative research is broadly welcomed Jagosh et al. (2012) Cargo & Mercer (2008).
Accountability depends on institutional memory. Agreements, promised return of findings, incidents involving disclosure, and decisions about secondary use should survive personnel turnover. Universities can require transparent logs of commitments and provide resources for local audit. The resulting documentation makes trust more than a subjective attitude: it connects expectations to observable fulfillment, dispute resolution, and continued access to evidence Israel et al. (2026) Guillemin & Gillam (2004).
Ethics in Practice: Conflict, Reflexivity, and Repair
Field research generates ethical decisions that a protocol cannot enumerate fully. An interview may reveal a participant's fear of a community leader, a focus group may expose identifiable personal histories, and an unexpected result may be used by officials to justify exclusion. Guillemin and Gillam's distinction between procedural ethics and ethics in practice provides a framework for examining these moments. Reflexivity involves documenting how investigators' positions, assumptions, and institutional incentives influence choices, and making important changes reviewable by affected parties Guillemin & Gillam (2004).
Practical reflexivity can include records of translation disagreement, reasons for excluding a quotation, changes in recruitment, and the handling of conflicts between participants and advisory bodies. An ethics diary is most useful when connected to escalation rules, supervision, and corrective authority. Confidentiality rules must shape what the diary itself contains, since notes about disputes can become sensitive data. Independent complaint routes matter because the researcher accused of misconduct should not be the only person evaluating the allegation Buchanan et al. (2007) TCPS 2 (2022).
Repair may involve correcting a misleading description, changing access permissions, compensating uncompensated labor, issuing a community-facing explanation, or stopping an injurious analysis. Some harms cannot be fully reversed after a dataset has circulated; preventing dissemination can therefore be a more meaningful remedy than retrospective apology. A mature ethics system records incidents and evaluates whether its governance structure was capable of detecting and responding to them Carroll et al. (2020) Hudson et al. (2020).
Knowledge Sovereignty and the Governance of Interpretation
Knowledge sovereignty concerns authority over the definition, custody, use, interpretation, and circulation of knowledge, particularly when information is embedded in the political, cultural, or relational life of a people. The term has its strongest articulated institutional foundations in Indigenous data sovereignty, which links information governance with collective self-determination. Kukutai and Taylor's edited volume examines rights relating to data about Indigenous peoples and territories. It establishes a conceptual and political research agenda that must be interpreted through the relevant people's actual laws and agreements Kukutai & Taylor (2016) Smith (2021).
Interpretive authority is as consequential as possession of a spreadsheet. Researchers can make technically accurate numerical statements while drawing ungrounded conclusions about a community's character, intentions, or capacity. Community specialists can question construct translations, sampling frames, missing histories, and categories imposed by outside agencies. Participatory review can reveal misclassification and overlooked mechanisms; it can also produce conflicts over interpretation. A scientifically responsible agreement specifies the procedure for disagreement, including whether alternative interpretations appear in the final report Carroll et al. (2020) TCPS 2 (2022).
The boundary of this dossier is community epistemic governance. Generic data privacy and individual consent are essential supporting areas but belong to wider research ethics. Knowledge sovereignty asks whose interests a knowledge system is built to serve and who has authority to determine permissible representation. These questions apply to interviews, ethnographies, psychometric datasets, historical archives, genetic material, audiovisual recordings, and trained predictive models. Distinct types of data create distinct rights and harms; a governance framework must name the actual subject matter rather than invoke sovereignty as a substitute for operational decisions Hudson et al. (2020) AIATSIS (2020).
Indigenous Research Ethics and Collective Self-Determination
Indigenous research ethics frameworks address responsibilities extending beyond the risk–benefit calculus of an isolated research participant. The AIATSIS Code of Ethics for Aboriginal and Torres Strait Islander Research, published in 2020, foregrounds Indigenous values, worldviews, and responsibilities. Canadian TCPS 2 Chapter 9 describes engagement with First Nations, Inuit, and Métis communities, attention to welfare of the collective, cultural protocols, agreements, and appropriate authority structures. These sources are standards situated in particular institutions and jurisdictions; their practical application requires local interpretation AIATSIS (2020) TCPS 2 (2022).
Research can affect sacred narratives, languages, genealogies, culturally restricted knowledge, cultural heritage, and community identity. Even materials publicly accessible in a database may have been obtained historically without authorization. Reuse therefore raises the ethics of provenance and contextual legitimacy, as well as access under current law. The governance of existing records is not settled solely by the visibility of a URL. Researchers can assess the originating community, conditions of initial collection, relevant custodians, intended audiences, and potential harms of recontextualization TCPS 2 (2022) Smith (2021).
Practical respect includes recognizing diversity among Indigenous peoples, across nations and within them. The First Nations OCAP principles are associated specifically with First Nations information governance in Canada and should not be applied as an invented universal Indigenous law. Consultation with relevant knowledge holders and competent authorities, informed by applicable guidance, establishes the appropriate practices. Where participants face competing pressures, individual autonomy and collective stewardship require explicit negotiation rather than a blanket formula FNIGC (n.d.) Hudson et al. (2020).
OCAP and Institutional Control of Research Information
The First Nations principles of OCAP stand for ownership, control, access, and possession. The First Nations Information Governance Centre presents them as standards governing collection, protection, use, and sharing of First Nations data. Ownership concerns the relationship between a community and its information; control concerns decision-making across stages of the information lifecycle; access concerns when and how information is obtainable; possession concerns practical custody and its relationship to governance. The four dimensions can separate in institutional research arrangements and need explicit contractual expression FNIGC (n.d.). OCAP® is a registered trademark of the First Nations Information Governance Centre (FNIGC), and use of the term should preserve that attribution FNIGC (n.d.).
A university might grant a community consultation rights without data custody, or share files while retaining unilateral authority to authorize secondary analysis. Neither arrangement alone establishes effective control. An agreement should identify who administers access credentials, approves transfers, maintains backups, authorizes derived datasets, sets retention schedules, and can suspend access after a breach. The contract should also explain how information remains available for legitimate scientific scrutiny under agreed conditions. This transforms broad commitments into assigned operational duties Carroll et al. (2020) TCPS 2 (2022).
OCAP governance raises questions about continuity after funding ends. Servers, archival collections, research assistants, and institutional positions may change; researchers can document succession, contact mechanisms, and the form in which data and metadata can be handed back. Where legal or contractual rules differ across parties, negotiation must occur before extraction and upload. A paper describing OCAP compliance should identify the specific practices established, rather than relying on the acronym alone as a certificate FNIGC (n.d.) Hudson et al. (2020).
CARE, FAIR, and the Limits of Universal Open Data
The CARE Principles for Indigenous Data Governance—Collective Benefit, Authority to Control, Responsibility, and Ethics—were articulated through an international Indigenous data sovereignty network and collaborators. Their emphasis on Indigenous rights and interests addresses important governance questions when data are shared or processed for secondary purposes. In parallel, the FAIR principles seek findable, accessible, interoperable, and reusable data to improve the infrastructure for scientific reproducibility and computational analysis. These frameworks address different dimensions and can be jointly considered where actual access policies allow Carroll et al. (2020) Wilkinson et al. (2016).
FAIR's accessibility requirement explicitly permits authentication and authorization protocols; it should not be mistaken for a requirement that every record be openly downloadable. Metadata can be publicly discoverable while the dataset remains access-controlled. CARE adds an account of who benefits and who may legitimately make decisions about Indigenous data. Practical coordination can use controlled repositories, tiered permissions, metadata restrictions for sensitive places, reciprocal agreements, and logs of analytic use. The appropriate design depends on data sensitivity, jurisdiction, and the agreements of the people whose information is involved Wilkinson et al. (2016) Carroll et al. (2020).
Data sharing decisions are empirical as well as ethical. Open datasets can strengthen checking and independent analysis; uncontrolled release can enable re-identification, stigma, extractive commercialization, or models that encode harmful categories. The challenge is to preserve the ability to contest scientific claims while managing harms from irreversible replication. Reproducibility may be supported by accessible protocols, approved code execution, protected auditing, and synthetic demonstrations, although none automatically substitutes for independent examination of actual records Narayanan & Shmatikov (2009) Hudson et al. (2020).
Privacy, Re-identification, and Derived Data
Community research datasets can identify people through combinations of quasi-identifiers, relationship structure, rare events, language, geography, and organizational roles. Narayanan and Shmatikov's work on de-anonymizing social networks demonstrated that structural information can undermine conventional assumptions that removing names ensures anonymity. A local study with few participants may face severe identification risk even when names are removed. Disclosure evaluation therefore needs an adversarial analysis of data linkage, outside information, and the scale of the population Narayanan & Shmatikov (2009).
Derived information deserves independent scrutiny. Geocoded measures, sentiment labels, diagnostic categories, embeddings, inferred group attributes, and model predictions may contain new risks and may reproduce sensitive source relationships. Agreements should state whether outputs count as controlled research products, how accuracy is assessed, and who can inspect training and inference records. Community names may remain inferable even after person-level identifiers are hidden. In genomic contexts, Hudson and collaborators discuss Indigenous rights and expectations that challenge unrestricted data access Hudson et al. (2020).
Security and confidentiality must be distinguished from ownership and permissible interpretation. Encryption limits unauthorized access, but it does not specify legitimate ownership, community review, or who may publish results. Conversely, a strong governance charter cannot repair a preventable breach of passwords or unsafe sharing of raw files. Audit requires both technical safeguards and institutional authority: data inventories, access records, review of outputs, breach response, third-party processor arrangements, and end-of-project disposition Carroll et al. (2020) TCPS 2 (2022).
Participatory Partnerships and the Evidence on Their Effects
Participatory research is often justified through the expectation that it improves relevance, recruitment, interpretation, implementation, and legitimacy. Those are distinct outcomes. Israel and colleagues' 1998 review described principles and recurring obstacles, while the 2026 update identified substantial development of the field, including a larger partnership literature and validated evaluation tools. These syntheses support the importance of partnership as a research program; they do not imply that any particular participatory label guarantees better health or more reliable inference Israel et al. (1998) Israel et al. (2026).
Jagosh and colleagues' realist review examined circumstances under which participatory research can generate benefits. Its context-sensitive explanatory approach is valuable because partnership arrangements are heterogeneous in intensity, institutional support, and purpose. Oetzel and colleagues' 2018 research tested relationships in a CBPR conceptual model and provides empirical evidence on associations among partnership processes and reported effects. Such analyses strengthen explanation by making mechanisms measurable but cannot settle all causal questions when exposure to collaboration is not randomly assigned Jagosh et al. (2012) Oetzel et al. (2018).
An evaluation should separate project participation from the outcomes participants care about. Increased attendance at meetings is an intermediate process indicator. Improved survey response, accurate interpretation, changes in institutional policy, reduced disparities, and durable community capacity are different endpoints with different time scales. Measuring only short-term perceptions of collaboration may inflate apparent success when tangible outcomes require years. A strong design predefines outcomes with partners and includes negative or disappointing findings Cargo & Mercer (2008) Wallerstein & Duran (2010).
Epistemic Quality, Scientific Independence, and Disagreement
Community interpretation can improve construct validity by revealing inadequate translations, misleading category boundaries, implausible causal pathways, and historically consequential omissions. Local partners often know what the investigator's measures fail to capture. The contribution is strongest when competing explanations are made explicit and can be tested with observations, qualitative evidence, or sensitivity analysis. A partner's lived knowledge merits careful examination and appropriate attribution; it does not render a factual claim immune to contradictory evidence Israel et al. (2026) Buchanan et al. (2007).
Buchanan, Miller, and Wallerstein analyzed ethical tensions between scientific rigor and participation in intervention research. Their discussion makes clear why governance has methodological consequences: different stakeholders may disagree over randomization, withholding a program, collecting sensitive variables, or reporting unfavorable outcomes. Scientific independence requires a defensible process for resolving conflict, documenting objections, and protecting the analysis from sponsor censorship. Community governance, similarly, should include safeguards against the capture of results by politically powerful local actors Buchanan et al. (2007).
A practical disagreement protocol specifies the underlying question, the evidence relevant to resolving it, the parties entitled to respond, and the disclosure needed if uncertainty remains. Interpretation panels may separate claims about data, causal inference, cultural meaning, and policy preference. A report can state that an estimate is statistically supported while acknowledging disagreement about its ethical significance. This keeps public debate grounded in identifiable evidence while preserving the authority of affected groups to interpret how research interventions bear on their lives TCPS 2 (2022) Guillemin & Gillam (2004).
Measurement of Partnership Governance and Accountability
Governance quality is a multidimensional construct. Candidate dimensions include agenda-setting influence, clarity of authority, representativeness of participation, fairness of compensation, quality of communication, transparent access decisions, response to complaints, dissemination agreements, and realized benefits. A composite score should be justified by its measurement model: high performance on communication cannot necessarily compensate for absent authority over data release. Researchers can report dimension-specific indicators alongside any total score Oetzel et al. (2018) Cargo & Mercer (2008).
Measurement validity requires attention to the referent of each item and the position of each respondent. An academic principal investigator, a paid community coordinator, and an unconsulted resident may give different accounts of the same arrangement. Confidential sampling of multiple roles, documentary verification of agreements, and direct observation of decision procedures can reduce dependence on a single reporting channel. Data should be disaggregated only where confidentiality and sample support permit, with explicit handling of missing or suppressed categories Jagosh et al. (2012) Israel et al. (2026).
Outcome validation benefits from longitudinal audit. A partnership that successfully negotiates an agreement at startup may fail to deliver accessible findings two years later. Researchers should follow the sequence from announced promise to assigned responsibility to documented action and affected-party assessment. An accountability indicator should record who evaluated fulfillment, who could challenge the evaluator, and what remedy followed failure. This makes an ethical obligation empirically inspectable without pretending that all normative questions reduce to survey numbers Guillemin & Gillam (2004) TCPS 2 (2022).
Benefit Sharing, Authorship, and the Return of Research Findings
Benefit sharing can involve access to results, equitable compensation, services, policy influence, institutional capacity, training, credited intellectual work, and permitted downstream income. Each benefit has a different beneficiary and timeline. A study may provide useful community reports without producing material change; another may generate policy recommendations whose adoption is controlled by public agencies. Research agreements should specify deliverables within the project's authority and distinguish realistic commitments from desired downstream effects Wallerstein & Duran (2010) Cargo & Mercer (2008).
Authorship decisions should recognize substantive contributions to conceptualization, analysis, interpretation, and writing under the applicable journal criteria, while also providing meaningful alternatives for acknowledging community knowledge and labor. Institutions may need translation, mentorship, or administrative support to enable genuine participation in publication. The choice between naming a contributor and maintaining confidentiality can itself require negotiation. A community advisory acknowledgment should never substitute for appropriate credit when the actual work warrants authorship TCPS 2 (2022) AIATSIS (2020).
Returning findings is an interpretive process rather than merely sending a finished PDF. Languages, formats, technical accessibility, and the ability to ask questions determine whether results are usable. Early review may identify misinterpretation, but the agreement must also protect legitimate reporting of unfavorable evidence. The strongest arrangement provides a specified period for response and correction, records unresolved differences, and communicates final outputs in forms usable by participants, practitioners, and affected institutions Israel et al. (2026) Carroll et al. (2020).
Cross-cultural Research, Translation, and Structural Inequality
Psychological constructs are shaped by the language, institutions, and conditions under which they are measured. Questions about autonomy, trust, household roles, collective decision-making, and psychological distress can operate differently across settings. Community involvement can identify culturally inappropriate response options and missing explanatory context. Yet a culturally familiar instrument still needs empirical evaluation of reliability, item functioning, and comparability if the study intends to compare populations. Good ethical consultation informs methodological validation; neither substitutes for the other Israel et al. (1998) Israel et al. (2026).
Research protocols should specify when interpretation is sensitive to migration status, disability, class, age, language proficiency, gender, and political vulnerability. These characteristics can influence who is invited, who can safely refuse, and how interview claims are interpreted. Translation teams need procedures for resolving semantic disagreements and for documenting whether a construct has been changed during adaptation. The risk of stigma may be magnified when a study labels a group deficient without measuring unequal access to resources and institutional constraints TCPS 2 (2022) Smith (2021).
Structural inequality also enters through scientific institutions. Researchers with stable funding may control a project in settings where local organizations have shorter contracts and less administrative capacity. Transparent budget decisions, local research training, and joint governance can partly rebalance this relationship. The empirical question is whether such arrangements produce meaningful authority and improved evidence, not whether participatory terminology appears in a grant application Cargo & Mercer (2008) Wallerstein & Duran (2010).
Digital Communities, Platforms, and Artificial Systems
Research involving digital communities complicates the distinction between accessible information and ethically reusable knowledge. Posts can be public to a platform's users while creators reasonably expect limited contextual circulation. Scraping can connect material across time, account identities, and geographic networks, creating new inference risks. Research agreements must therefore investigate original audience, applicable law, platform rules, vulnerability to linkage, and the consequences of quoting searchable language. A public source is not automatically a justification for unrestricted republication of sensitive collective characteristics Narayanan & Shmatikov (2009) Guillemin & Gillam (2004).
Artificial systems extend the data lifecycle from records to learned representations, automated classification, and generated interpretation. Training models on interviews or cultural archives can alter who accesses knowledge, the scale at which it is recombined, and whether sensitive information becomes recoverable through outputs. Community research ethics can evaluate whether the training purpose was authorized, whether restrictions can be enforced against downstream processors, and how errors affecting communities are detected and corrected. These are research governance questions, rather than evidence that artificial systems possess any particular psychological state Carroll et al. (2020) Hudson et al. (2020).
At the horizon of the Artificial Era, human–artificial research relations introduce new delegated agents into analysis, drafting, and data administration. The relevant scientific tests concern observable control structures: accountable human and organizational principals, traceable model inputs and outputs, documented review authority, and mechanisms to contest decisions. The philosophical language of Artificial Sapiens can organize questions about nonbiological reasoning and agency, while empirical claims about system competence, reliability, and effects require separate measurement. Communities must know when substantive interpretations are prepared by automated systems and who remains responsible for publication Wilkinson et al. (2016) Carroll et al. (2020).
Governance Agreements as Testable Research Infrastructure
A robust research agreement can be modeled as a structured account of responsibilities rather than an expression of goodwill. It should specify the problem under study, legitimate parties, protocol approval, participant safeguards, ownership and custodianship, access levels, permitted reuse, analytic participation, interpretation review, dissemination, credit, financial arrangements, complaint procedures, and project exit. Different provisions will need different signatories and different oversight. The agreement's actual function can be tested by inspecting decision records, resource transfers, and documented responses to conflict TCPS 2 (2022) FNIGC (n.d.).
Governance also requires a schedule for change. A new collaborator, repository migration, security incident, research question, or commercial partnership can alter the conditions under which information was originally provided. An amendment mechanism should identify the threshold for renewed review and who can suspend work while a dispute is considered. An external reviewer or mutually chosen mediator can reduce the risk that only the strongest partner interprets the agreement. These are proposed design principles drawn from the documented limitations of one-time consent and partnership power disparities Guillemin & Gillam (2004) Buchanan et al. (2007).
Scientific accessibility can be built into agreements from the start through preplanned analyses, transparent code and materials, controlled audit environments, and review of restricted variables. A demand for reproducibility should identify what level of independent checking is necessary and what alternative paths are available when public data release would create significant harm. A demand for control should identify the specific legitimate interests protected and how accurate findings can still be questioned. These design tests connect FAIR-oriented scientific quality with CARE-oriented responsibility Wilkinson et al. (2016) Carroll et al. (2020).
Evaluation of Harm, Redress, and Long-term Accountability
Community harms may include stigmatizing public interpretation, unauthorized reuse of culturally restricted materials, exposure of vulnerable individuals through small-cell statistics, discriminatory deployment of predictive tools, and disruption of local trust. The likelihood and severity of harms vary by audience, circulation channel, local power relations, and the reversibility of disclosure. Risk assessment should consider both immediate interview burdens and foreseeable effects after findings leave the research partnership. Precaution becomes most compelling where disclosure is difficult to reverse Hudson et al. (2020) TCPS 2 (2022).
A redress framework needs concrete procedures: accessible complaints; confidential handling; independent investigation; thresholds for pausing dissemination; a correction protocol; transparent communication about breaches; and a means to seek restitution or other appropriate remedy. Responsibility should be assigned across universities, funders, publishers, repository administrators, and third-party data processors. Written protocols cannot guarantee justice, but their presence and implementation can be independently examined. Ethical success includes an institution's capacity to acknowledge error and change its practice Guillemin & Gillam (2004) National Commission (1979).
Long-term accountability is especially relevant to cumulative and reusable data. A repository may outlive the organization that negotiated the original agreement; a derivative model can persist after individual files are withdrawn; a public claim about a community may recur in later scholarship. Stewardship plans should specify successor authorities, review cycles, access logging, retention and deletion terms, and the management of downstream products. Follow-up studies should examine whether benefits and harms remain concentrated in the same groups over time Carroll et al. (2020) Kukutai & Taylor (2016).
Researcher Positionality and the Production of Evidence
The status of an investigator as a community insider, outsider, former participant, clinician, institutionally affiliated professional, or political actor shapes opportunities to recruit, observe, and interpret. Positionality is a factual feature of access and accountability: it affects what people reveal, what they expect from research, and whose account receives institutional credibility. Researchers can describe relevant roles, conflicts of interest, previous organizational relationships, and the resulting implications for methods. Reflexive documentation becomes scientifically useful when it identifies how these relations changed sampling or interpretation, rather than presenting personal biography as proof of valid findings Guillemin & Gillam (2004) Smith (2021).
Insider research can provide language fluency and recognized obligations, but proximity can constrain the expression of criticism, especially within hierarchical communities. Outsider researchers can sometimes protect independent scrutiny yet misunderstand situated practices. Strong designs can distribute roles, use independent interviews, encourage participant corrections, and document when researchers disagree. Credibility rests on clear procedures and triangulated observations, with careful attention to the confidentiality of people who contest community leaders. Participatory work must therefore be evaluated through actual practices, not a categorical preference for one investigator identity Israel et al. (1998) Cargo & Mercer (2008).
Epistemic reflexivity can be operationalized through analytic memos recording the source of an interpretation, the evidence that might falsify it, how disputed terms were translated, and the decisions reached after partner feedback. Auditors can compare successive drafts against participant records without requiring public access to sensitive transcripts. Disclosure of limitations then becomes a substantive account of what the available evidence permits. This is particularly important where researchers derive group-level psychological claims from small or strategically selected samples Guillemin & Gillam (2004) TCPS 2 (2022).
Community Trials, Intervention Consent, and Cluster-Level Risks
Community interventions may be delivered through schools, neighborhoods, clinics, social networks, or local organizations. The entity assigned to an intervention may therefore differ from the individuals who experience it. Cluster randomization raises questions of authorization by gatekeepers, individual recruitment, spillovers across boundaries, and whether withholding an intervention creates ethically relevant inequity. Scientific designs need to identify allocation units, plausible interference, outcomes, and participants who bear burdens, while ethical review considers which decisions a collective body can legitimately make Buchanan et al. (2007) National Commission (1979).
Programs may modify a local environment through communications, services, incentives, or changes in institutional rules. An informed resident may still be indirectly affected because the intervention changes surrounding opportunities. Researchers should map exposure beyond direct participants and examine whether outcomes for unmeasured groups might differ. If evaluation compares neighboring communities, spillover and unequal baseline services threaten causal interpretation. A protocol that reports only individual participants' consent and responses may miss the full field of potential effects Wallerstein & Duran (2010) Jagosh et al. (2012).
Public accountability becomes important when an intervention is adopted as policy. Implementation authorities, researchers, service providers, and residents may have divergent expectations about evidence and benefit. A study must distinguish the hypothesis it tested from a broader policy claim and report who made decisions about adaptation. Participatory evaluation can increase visibility of unintended consequences but still needs systematic outcome assessment. Neither scientific randomization nor community endorsement alone is sufficient to justify an intervention's risks and intended use Buchanan et al. (2007) Israel et al. (2026).
Institutional Review Boards and Community Review Bodies
Institutional review boards and research ethics committees assess human-participant protections within their applicable regulatory and institutional mandates. Their review usually emphasizes risks, consent procedures, privacy, selection of participants, and protection of vulnerable persons. Community review bodies can bring knowledge of collective consequences, local authority, appropriate cultural protocols, and histories of research that institutional reviewers may not possess. A coherent process identifies the authority and scope of each body and the procedure for responding when recommendations conflict HHS OHRP (n.d.) TCPS 2 (2022).
Multiple review processes can increase accountability but may also create delay and administrative burdens for organizations with limited resources. A risk-based design can coordinate documentation, recognize existing review capacity, and clarify which approval is legally required versus requested by agreement. Researchers should budget for deliberation and translation before recruitment begins, because deadlines do not extinguish legitimate obligations. Community reviewers themselves need independence and transparent conflict-of-interest rules where their organizations receive project funding AIATSIS (2020) TCPS 2 (2022).
Review should continue when a project's scientific identity changes. Replacing interviews with passive digital tracking, adding genomic variables, or publishing geographic predictions can transform the risk profile. Amendments should reach the bodies responsible for the newly implicated interests, with meaningful time for review. An auditable record distinguishes the initial conditions of approval from later expansions, making it possible to determine whether the actual project remained within its authorization Guillemin & Gillam (2004) Carroll et al. (2020).
Knowledge Repositories, Data Access Committees, and Auditability
Repositories are institutions that enforce or frustrate research agreements through everyday technical actions. A repository may publish descriptive metadata while limiting access to person-level records, require a data-use agreement, prohibit certain derived outputs, and retain logs of who accessed restricted material. Governance quality depends on whether those constraints correspond to rights that affected communities recognize and whether they survive changes in staff or vendors. CARE and FAIR offer complementary vocabulary for specifying both stewardship responsibilities and scientific reuse conditions Carroll et al. (2020) Wilkinson et al. (2016).
Access committees should have defined criteria, competent representation, and a procedure for considering scientific merit, consent scope, collective interests, commercial use, and foreseeable risk. They should also state the grounds for refusal and options for revision or appeal. Delegated control can become arbitrary if review is opaque; reciprocal transparency can reduce that risk without exposing confidential deliberations. Practical evaluations should compare stated rules with actual grants of access and secondary publications FNIGC (n.d.) Hudson et al. (2020).
An effective audit examines not just downloads but transformations and transfers. A dataset can be copied into a new institution, converted to feature vectors, linked to another registry, or used to train a predictive model. Tracking provenance, authorized purposes, derivation history, and output approvals enables attribution of responsibility. Some obligations are legal and others contractual or ethical; documented authority is needed before treating any technical restriction as universally binding. Long-term governance depends on enforceable agreements and feasible records of downstream use Kukutai & Taylor (2016) Carroll et al. (2020).
Credible Knowledge Translation and Public Communication
The ethical consequences of research do not end when a manuscript is accepted. A scientific description of differences between communities can become a headline, administrative category, or justification for resource allocation. The audience may ignore uncertainty, selection effects, and the narrow operational definition originally used. Authors and institutions can reduce misinterpretation by publishing clear descriptions of what was measured, who was included, alternative explanations, and the limits of generalization. Community review can identify damaging ambiguities before dissemination, while the published record must retain accurate and contestable evidence Israel et al. (2026) National Commission (1979).
Community-facing communication involves choices about medium, language, accessibility, timing, and security. A public meeting may be inappropriate for disclosing sensitive individual histories; an encrypted briefing may reach accredited custodians but exclude residents who need accessible interpretation. Formats should be matched to legitimate audiences, and agreements should identify where participants may seek clarification or challenge an error. These details transform the promise to 'return results' into a process that can be measured Cargo & Mercer (2008) Wallerstein & Duran (2010).
Researchers can distinguish confirmed findings from preliminary explorations, causal interpretations from correlations, and evidence from policy preferences. Where disagreement persists, the publication may describe the distinct positions and the supporting evidence without presenting consensus that was never obtained. Postpublication corrections and withdrawal of an unsafe ancillary dataset should be possible through predetermined channels. Communication quality is thus a continuing test of accountability and scientific integrity Guillemin & Gillam (2004) TCPS 2 (2022).
Training, Research Capacity, and Intergenerational Stewardship
Capacity development is frequently treated as a benefit of community partnerships, but its substance varies. It may include training in interviewing, statistics, governance, budgeting, data security, qualitative analysis, or negotiation with funders. Evaluation should examine whether participants acquire usable skills, retain equipment and institutional access, and can initiate independent research after the original project ends. Counting workshop attendance is a limited proxy for durable research authority. The distinction matters because formal participation can persist without local control of resources Israel et al. (2026) Oetzel et al. (2018).
Intergenerational stewardship is relevant when research documents language, place, genealogies, cultural practices, and collective histories. Knowledge holders may hold obligations to future members as well as to current participants. Archival choices about permanent storage, copying, culturally restricted access, and withdrawal rights therefore have temporal consequences beyond a grant cycle. Researchers should negotiate preservation priorities and succession arrangements with competent custodians before deciding that perpetual open availability is automatically a public benefit Kukutai & Taylor (2016) AIATSIS (2020).
Practical questions include who will maintain a database after academic personnel leave, which institution can authorize future uses, how corrections to metadata are recorded, and whether younger generations have channels for legitimate review. These are context-specific decisions rather than assumptions about a single collective will. The value of intergenerational planning can be assessed through continuity of access, functioning decision bodies, archived provenance, and evidence that promised knowledge returns actually occurred FNIGC (n.d.) Carroll et al. (2020).
A Research Agenda for Validating Ethical Governance
A useful program of future research would compare alternative forms of partnership authority without reducing them to a single participation score. Candidate observational designs include longitudinal studies of research agreements, multi-stakeholder surveys, linked audits of access requests, and case comparisons where institutions face similar questions but organize decisions differently. Researchers would preregister key hypotheses when feasible, record contextual differences, and separate process indicators from scientific and community outcomes. Qualitative process tracing could identify how decisions produce particular benefits or failures Oetzel et al. (2018) Jagosh et al. (2012).
Construct validation should test whether measures of trust, procedural justice, representativeness, and accountability have equivalent meanings across settings. Researchers can collect reports from academic partners, local organizers, recruited participants, and residents whose interests are affected but who did not participate. Administrative records offer independent evidence about compensation, decision rights, complaints, and access permissions. Ethical obligations may vary normatively even where measurement characteristics are similar; cross-setting comparisons should make those differences visible Cargo & Mercer (2008) TCPS 2 (2022).
The strongest explanatory contribution would identify conditions under which a governance arrangement improves inference, distributes benefits, and reduces foreseeable harm, while documenting counterexamples. Such studies could assess trajectories of trust, rates of unauthorized data reuse, reproducibility under controlled access, and the correction of inaccurate community-level claims. Results should be interpreted with explicit causal assumptions and risks of selection into partnership. This agenda turns ethics into an empirically accessible domain of institutional practice without claiming that ethical justification itself is exhausted by measured outcomes Israel et al. (2026) Carroll et al. (2020).
Scientific Disagreements and Unresolved Problems
Several questions remain empirically open. Which dimensions of decision sharing predict independent gains in data quality or outcomes? Under what conditions does intensive participation improve recruitment, reduce selection bias, or increase retention? When might collaborative governance favor the preferences of more powerful community intermediaries? How can benefits be compared across studies with radically different purposes and timescales? Existing model tests and reviews indicate promising processes, but heterogeneous designs and self-selection constrain universal causal conclusions Oetzel et al. (2018) Jagosh et al. (2012) Israel et al. (2026).
Knowledge governance also generates unresolved normative conflicts. A future-use proposal can advance valuable public knowledge while contravening the expectations of knowledge custodians. A community body may request suppression of unfavorable findings to avoid stigma. An individual may wish to disclose their own story despite collective restrictions. Legitimate authority is domain-specific, and independent scientific reporting must be balanced through justified processes rather than an automatic priority rule. The research task is to document how different arrangements protect autonomy, stewardship, truthfulness, and justice in concrete conditions TCPS 2 (2022) Buchanan et al. (2007).
New technologies sharpen these issues. Linkage and large language models make it increasingly difficult to treat a released dataset as permanently anonymous or to assume that downstream uses will remain close to the original purpose. Research is needed on enforceable machine-readable permissions, community oversight of model outputs, trustworthy restricted-data validation, transnational jurisdiction, and the longevity of consent across generations. These are proposed priorities arising from identified technological and governance problems, not established evidence of the efficacy of any single solution Narayanan & Shmatikov (2009) Hudson et al. (2020) Carroll et al. (2020).
Conclusion
Community research ethics is the study and governance of duties created when scientific knowledge is produced within relationships among persons, groups, researchers, and institutions. Its distinctive problem is the distribution of epistemic authority across the complete research lifecycle. Participation is meaningful when communities can influence decisions affecting them, while individuals retain protections relevant to their own choices and welfare. Collective stewardship requires attention to actual histories, constituencies, political standing, and cultural knowledge systems National Commission (1979) TCPS 2 (2022) Israel et al. (1998).
The literature supports several disciplined conclusions. Research partnerships create identifiable opportunities to improve relevance, interpretation, reciprocal knowledge translation, and institutional responsiveness. Realist synthesis, conceptual models, and partnership studies offer evidence about plausible mechanisms, yet causal effects vary by setting and outcome and remain difficult to isolate. Indigenous data governance frameworks demonstrate concrete approaches to rights, control, access, responsibility, and benefit, while security research explains why uncontrolled disclosure can threaten people and communities even after conventional de-identification Jagosh et al. (2012) Oetzel et al. (2018) Carroll et al. (2020) Narayanan & Shmatikov (2009).
The practical standard emerging from this synthesis is accountable, documented co-governance: clear research objects, justified representation, voluntary individual participation, context-sensitive rights over knowledge, transparent handling of disagreement, rigorous methods, and remedies that survive the life of a grant. Scientific credibility and social responsibility are jointly tested through evidence, records, and consequences. This monograph establishes a research map for examining these relationships in community psychology and for studying their transformation as artificial systems increasingly participate in scientific knowledge production Israel et al. (2026) Guillemin & Gillam (2004) Wilkinson et al. (2016).
References
- Australian Institute of Aboriginal and Torres Strait Islander Studies. (2020). AIATSIS code of ethics for Aboriginal and Torres Strait Islander research. AIATSIS. https://aiatsis.gov.au/research/ethical-research/code-ethics
- Buchanan, D. R., Miller, F. G., & Wallerstein, N. (2007). Ethical issues in community-based participatory research: Balancing rigorous research with community participation in community intervention studies. Progress in Community Health Partnerships: Research, Education, and Action, 1(2), 153–160. https://doi.org/10.1353/cpr.2007.0006
- Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada, & Social Sciences and Humanities Research Council of Canada. (2022). Tri-Council policy statement: Ethical conduct for research involving humans (TCPS 2), Chapter 9: Research involving the First Nations, Inuit and Métis peoples of Canada. Government of Canada. https://ethique.gc.ca/eng/tcps2-eptc2_2022_chapter9-chapitre9.html
- Cargo, M., & Mercer, S. L. (2008). The value and challenges of participatory research: Strengthening its practice. Annual Review of Public Health, 29, 325–350. https://doi.org/10.1146/annurev.publhealth.29.091307.083824
- Carroll, S. R., Garba, I., Figueroa-Rodríguez, O. L., Holbrook, J., Lovett, R., Materechera, S., Parsons, M., Raseroka, K., Rodriguez-Lonebear, D., Rowe, R., Sara, R., Walker, J. D., Anderson, J., & Hudson, M. (2020). The CARE principles for Indigenous data governance. Data Science Journal, 19, Article 43. https://doi.org/10.5334/dsj-2020-043
- First Nations Information Governance Centre. (n.d.). The First Nations principles of OCAP®: Ownership, control, access, and possession. First Nations Information Governance Centre. https://fnigc.ca/ocap-training/
- Guillemin, M., & Gillam, L. (2004). Ethics, reflexivity, and “ethically important moments” in research. Qualitative Inquiry, 10(2), 261–280. https://doi.org/10.1177/1077800403262360
- Hudson, M., Garrison, N. A., Sterling, R., Caron, N. R., Fox, K., Yracheta, J., Anderson, J., Wilcox, P., Arbour, L., Brown, A., Taualii, M., Kukutai, T., Haring, R., Te Aika, B., Baynam, G. S., Dearden, P. K., Chagné, D., Malhi, R. S., Garba, I., . . . Carroll, S. R. (2020). Rights, interests and expectations: Indigenous perspectives on unrestricted access to genomic data. Nature Reviews Genetics, 21, 377–384. https://doi.org/10.1038/s41576-020-0228-x
- Israel, B. A., Schulz, A. J., Becker, A. B., Reyes, C. L., Parker, E. A., & Reyes, A. G. (2026). Community-based participatory research: Evolution and significant developments. Annual Review of Public Health, 47, 135–157. https://doi.org/10.1146/annurev-publhealth-082224-022223
- Israel, B. A., Schulz, A. J., Parker, E. A., & Becker, A. B. (1998). Review of community-based research: Assessing partnership approaches to improve public health. Annual Review of Public Health, 19, 173–202. https://doi.org/10.1146/annurev.publhealth.19.1.173
- Jagosh, J., Macaulay, A. C., Pluye, P., Salsberg, J., Bush, P. L., Henderson, J., Sirett, E., Wong, G., Cargo, M., Herbert, C. P., Seifer, S. D., Green, L. W., & Greenhalgh, T. (2012). Uncovering the benefits of participatory research: Implications of a realist review for health research and practice. Milbank Quarterly, 90(2), 311–346. https://doi.org/10.1111/j.1468-0009.2012.00665.x
- Kukutai, T., & Taylor, J. (Eds.). (2016). Indigenous data sovereignty: Toward an agenda. ANU Press. https://doi.org/10.22459/CAEPR38.11.2016
- Narayanan, A., & Shmatikov, V. (2009). De-anonymizing social networks. In 2009 30th IEEE Symposium on Security and Privacy (pp. 173–187). IEEE. https://doi.org/10.1109/SP.2009.22
- National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (1979). The Belmont report: Ethical principles and guidelines for the protection of human subjects of research. U.S. Department of Health and Human Services. https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/read-the-belmont-report/index.html
- Oetzel, J. G., Wallerstein, N., Duran, B., Sanchez-Youngman, S., Nguyen, T., Woo, K., Wang, J., Schulz, A., Kaholokula, J. K., Israel, B., & Alegría, M. (2018). Impact of participatory health research: A test of the community-based participatory research conceptual model. BioMed Research International, 2018, Article 7281405. https://doi.org/10.1155/2018/7281405
- Smith, L. T. (2021). Decolonizing methodologies: Research and Indigenous peoples (3rd ed.). Zed Books. https://www.bloomsbury.com/us/decolonizing-methodologies-9781786998125/
- U.S. Department of Health and Human Services, Office for Human Research Protections. (n.d.). 45 CFR 46: Protection of human subjects. U.S. Department of Health and Human Services. https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/index.html
- Wallerstein, N., & Duran, B. (2010). Community-based participatory research contributions to intervention research: The intersection of science and practice to improve health equity. American Journal of Public Health, 100(Suppl. 1), S40–S46. https://doi.org/10.2105/AJPH.2009.184036
- Wilkinson, M. D., Dumontier, M., Aalbersberg, I. J. J., Appleton, G., Axton, M., Baak, A., Blomberg, N., Boiten, J.-W., da Silva Santos, L. B., Bourne, P. E., Bouwman, J., Brookes, A. J., Clark, T., Crosas, M., Dillo, I., Dumon, O., Edmunds, S., Evelo, C. T., Finkers, R., . . . Mons, B. (2016). The FAIR guiding principles for scientific data management and stewardship. Scientific Data, 3, Article 160018. https://doi.org/10.1038/sdata.2016.18