RESEARCH DOSSIER
Community-Based Participatory Research
Abstract. Community-based participatory research (CBPR) organizes scientific investigation through sustained involvement of affected communities in the definition, design, conduct, interpretation, governance, and uses of research. This fundamental monograph examines CBPR as an object of psychological and methodological inquiry, reconstructing its historical foundations and distinguishing it from other forms of community engagement. It analyzes power sharing, representation, trust, reciprocity, collaborative knowledge production, sampling, qualitative inquiry, experimental and quasi-experimental designs, and the measurement of partnership processes. The evidence synthesis considers original participatory studies, the CBPR conceptual model, a multisite empirical test, realist and systematic reviews, mental-health outcomes, and the conditions under which policy and structural change may follow. Ethical inquiry addresses individual and collective consequences, data stewardship, unequal burdens, and long-term accountability. The analysis separates observed partnership processes from associations with intervention outcomes and from causal claims about participation. It identifies priorities for longitudinal research, psychometric validity, causal designs, community representation, and digitally mediated participation within the Artificial Era.
Keywords: community-based participatory research; CBPR; community psychology; participatory action research; community engagement; shared governance; participatory research methods; partnership synergy; health equity; knowledge co-production; mixed methods; research ethics
The Scientific Object of Community-Based Participatory Research
Community-based participatory research (CBPR) is a sustained mode of scientific inquiry in which community stakeholders and researchers share consequential responsibility for defining a problem, constructing evidence, interpreting findings, and determining the uses of knowledge. The investigated object of this monograph is the participatory organization of research itself: who defines questions, whose observations count, how decisions are negotiated, how claims are warranted, and how findings acquire consequences in the social settings being studied. Community participation is therefore an investigable property of a research process rather than a descriptive characteristic of its geographic location. The foundational account of Israel et al. (1998) links partnership to attention to social and structural determinants; the later synthesis Israel et al. (2026) describes the evolution of principles, involvement continua, evaluation, and structural change. A study conducted in a neighborhood may contain no CBPR if the people concerned do not possess meaningful influence over its inquiry.
The unit of analysis changes with the research question. An individual participant can report experience; an organization can control data collection; a coalition can distribute resources and decision rights; a community can influence which outcomes count as useful; a policy system can determine whether the new evidence changes practice. CBPR studies often combine all these levels, but evidence from one level must be connected explicitly to another. This dossier examines both the credibility of empirical claims and the governance arrangements that shape their production. It distinguishes a participatory partnership from a single intervention, an advisory meeting, a public consultation, and a community service program. Each may be connected to CBPR, while the defining scientific question concerns distributed authority over research decisions. The distinction permits comparison between genuine partnership and the appearance of participation.
Six questions organize the synthesis. What establishes the community and the legitimacy of its representatives? Which historical and conceptual traditions shaped participatory inquiry? By what mechanisms might power sharing improve relevance, recruitment, interpretation, capacity, or policy uptake? Which designs can make causal and descriptive claims credible? What do comparative studies and systematic reviews actually establish about outcomes? Finally, which ethical and institutional arrangements sustain accountability after a grant ends? The research is a critical narrative synthesis, using identifiable original studies and published reviews rather than claiming an exhaustive systematic search. It evaluates arguments at the level supported by the consulted article or abstract and treats participatory legitimacy, measurement validity, intervention efficacy, and institutional change as separate outcomes Cargo & Mercer (2008) Jagosh et al. (2012).
Origins, Traditions, and Historical Development
The genealogy of CBPR crosses social psychology, action research, community development, health promotion, and traditions of emancipatory inquiry. Kurt Lewin's action-research program in the 1940s made systematic investigation relevant to planned collective change, while later participatory practices developed in multiple settings beyond North American institutions. Accounts of the northern pragmatic tradition and southern emancipatory traditions help explain why practical problem solving and transformation of unequal power relations coexist within the field Collins et al. (2018). Community partners have also developed their own knowledge practices independently of academic naming. Historical reconstruction therefore concerns intersecting traditions, institutional relationships, and the ways an academic category came to recognize activities that often predated its formation.
The published 1998 review by Israel, Schulz, Parker, and Becker provided an influential synthesis of community-based research principles, partnership conditions, and the relation between research and action Israel et al. (1998). Macaulay and colleagues' 1999 account emphasized collaboration between professional investigators and communities in primary care and Indigenous contexts Macaulay et al. (1999). The Agency for Healthcare Research and Quality's 2004 evidence assessment marked an important attempt to evaluate participatory research as an evidence-generating practice rather than treating it as an exclusively normative aspiration Viswanathan et al. (2004). Cargo and Mercer subsequently formulated an integrative practice framework connecting partnership development, research design, and evaluation Cargo & Mercer (2008). These texts form a historical sequence of conceptual clarification, methodological inquiry, and institutional legitimacy.
Later research shifted from describing principles to identifying partnership processes, building evaluative instruments, and explaining health and policy outcomes. The realist synthesis of Jagosh et al. (2012) investigated how participatory partnership produces effects in particular contexts. Belone et al. (2016) asked community partners to assess the face validity of a conceptual model; Oetzel et al. (2018) investigated its hypothesized pathways with survey and case-study evidence. Israel et al. (2026) revisits this evolving field over nearly three decades and highlights greater institutional capacity and validated partnership evaluation tools. The movement is historically important because a mature research program must permit questions about its own claims to be tested, including the possibility that participatory processes sometimes fail, redistribute burdens unfairly, or provide no additional benefit on a given outcome.
Conceptual Foundations and Neighboring Traditions
Participatory action research, community-engaged research, citizen science, co-production, public involvement, and CBPR occupy overlapping methodological territories. Their labels do not guarantee identical distribution of authority. Participatory action research traditionally emphasizes iterative cycles of inquiry and action; community-engaged research encompasses a wider spectrum of consultation and collaboration; co-production stresses joint development of knowledge or services; citizen science may involve large distributed contributions to scientific tasks. CBPR is identified here by the breadth, continuity, and consequence of community participation across the research life cycle. A useful classification describes concrete decision rights rather than attempting to settle terminology by dictionary decree Israel et al. (2026) Cargo & Mercer (2008).
Community is a relational and institutional concept requiring operational definition. It may refer to shared residence, social identity, service use, common experience of discrimination, organizational membership, or a network sustained across distance. Those membership criteria can overlap and change during a project. Community representatives therefore need an explicit warrant for their role and a procedure for accountability to people who are absent from partnership meetings. The assumption that one spokesperson can articulate a single community interest conceals inequalities of age, gender, class, language, citizenship, ability, and organizational influence. Meaningful participation requires attention both to who has a seat and to who can contest the decisions made in their name Macaulay et al. (1999) Israel et al. (2026).
The epistemological basis of CBPR is that knowledge about lived circumstances, organizational routines, environmental constraints, and practical feasibility can improve the specification of research problems and the interpretation of findings. This proposition is testable: a community-informed instrument may show stronger cultural comprehension, a recruitment protocol may reduce missed groups, or a co-designed intervention may be more acceptable. Such possibilities do not abolish standards of measurement, internal validity, or transparent inference. Partnership can introduce additional measurement error if participants feel pressure to provide favorable answers, while academic investigators can bias results through funding or publication incentives. The scientific task is to identify both sources of knowledge and conditions under which they support credible, reproducible findings Israel et al. (1998) Lasker et al. (2001).
The Research Life Cycle and the Location of Authority
Participatory quality can be assessed across linked phases: problem formulation, theory selection, sampling, protocol design, measurement, recruitment, data stewardship, analysis, interpretation, communication, and action. A partnership may share authority extensively during early planning and relinquish it at the analytical stage; another may include community researchers in data analysis but leave its research objectives fixed by a grant. Both arrangements disclose meaningful but different forms of participation. The relevant evidence is procedural and documentary: governance agreements, recorded decisions, budget authority, approved protocol changes, analytic workshops, dissemination choices, and the actual consequences of disagreement. A claim of full participation needs phase-specific evidence Israel et al. (2026) Cargo & Mercer (2008).
An initial needs assessment sets the trajectory of the project. Residents may identify problems that administrative statistics underestimate; practitioners may recognize logistical constraints in routine services; investigators may identify measurement weaknesses or previously tested explanations. Jointly selecting the research question requires criteria by which competing priorities are judged, particularly when urgency, feasibility, scientific novelty, and expected community benefit diverge. The resulting scope should document which priorities were rejected and why. During design, participation can alter language, time demands, compensation, and access to otherwise invisible networks. In analysis, it can clarify social meaning and expose implausible interpretations. Each contribution becomes strongest when its evidentiary role is specified and recorded.
Power over findings persists after data collection. The right to review results is different from the right to suppress unfavorable results; the right to protect identifiable community information is different from suppressing warranted criticism of institutions. Publication and dissemination agreements must delineate these cases before results become contentious. Joint authorship, plain-language reports, local presentations, policy briefs, and accessible data products represent different modes of accountability. For a scientifically credible CBPR project, decisions over privacy, interpretation, and use should remain visible to independent scrutiny, while communities retain appropriate control over sensitive information and planned uses. Israel et al. (2010) provides a concrete account of policy advocacy within a partnership; Wallerstein & Duran (2010) connects intervention research with mutual benefit and translation.
Power Sharing, Representation, and Partnership Governance
Power sharing is a set of observable institutional arrangements. It comprises the allocation of voting authority, protocol approval, data access, staffing, control of expenditures, and influence over the interpretation and use of results. Equal rhetorical recognition does not establish equal capacity to exercise these rights. Community groups may depend economically on the institution whose practices they are studying; researchers may hold exclusive authority over grant reports and ethics submissions; established organizations may monopolize invitations. A serious participation audit traces resources and decision procedures, identifies who can amend a plan, and records how conflicts are adjudicated. Structural power, organizational privilege, and the history of research relationships form the context in which partnership processes operate Oetzel et al. (2018) Belone et al. (2016).
Representation is a persistent scientific and political challenge. An advisory board may include respected local leaders but miss groups with precarious housing, irregular work, linguistic isolation, or hostile histories with local institutions. Snowball recruitment can deepen the influence of already connected participants, while a random household sample can omit people whose membership is social rather than territorial. Partnerships can combine formal elections, rotating seats, open assemblies, purposive recruitment, and feedback channels, evaluating the limitations of each arrangement. The legitimacy of a representative is therefore an empirical claim about responsiveness, inclusion, and accountability. Investigators must examine how dissent moves through the partnership and how materially affected people may challenge a misleading interpretation.
Governance must also survive asymmetries in expertise. Statistical modeling, law, financial management, and regulatory compliance require specialized knowledge, but specialization need not confer unchecked authority over normative priorities. One workable division assigns researchers responsibility for explaining technical constraints and validating analytic claims, and shared bodies responsibility for evaluating whether questions and uses remain consistent with agreed purposes. Joint vetoes or review procedures need clear conditions and documented remedies. The partnership model is stronger when it allows qualified disagreement and transparent escalation than when it treats harmony as the primary outcome. Lasker et al. (2001) framed the problem as collaborative advantage; Israel et al. (2026) emphasizes equitable partnerships and the long path from engagement to structural change.
Trust, Reciprocity, Conflict, and Psychological Processes
Trust develops through repeated evidence of competence, reliability, fairness, and care in high-stakes decisions. Communities with histories of extraction or misrepresentation may interpret research invitations through institutional memory rather than the present investigator's intentions. Such responses are rationally related to the risk profile of research participation. Researchers, meanwhile, may worry about delayed decisions, selective evidence, or unstable organizational mandates. Trust should be measured at multiple levels: confidence in particular people, confidence in governance procedures, and confidence in the institution's future conduct. The conceptual model examined by Belone et al. (2016) identified trust development, capacity, mutual learning, and power dynamics as cross-cutting constructs, based on consultation with six partnerships.
Reciprocity means that the burdens and benefits of inquiry are distributed according to an explicitly negotiated arrangement. Community researchers may provide cultural interpretation, recruitment labor, meeting facilitation, and public advocacy; meaningful compensation, training, authorship, and control of outputs recognize these contributions. Volunteer rhetoric can disguise unpaid labor, especially where contributors already carry economic or caregiving burdens. Reciprocity is better assessed by who receives resources and usable knowledge over time than by partner satisfaction alone. Researchers should record changes in participation, turnover, financial transfers, learning opportunities, and opportunities to withdraw without penalty. In this way, the moral language of equity becomes an empirical question about the functioning of institutions.
Productive disagreement can be an information source. Residents may reject a questionnaire category, service providers may challenge its feasibility, and academics may question an inference offered by a coalition. Each conflict can reveal hidden assumptions, but prolonged unaddressed conflict can damage relationships and compromise the study. Jagosh et al. (2012) identified negotiated conflict among the mechanisms through which participatory partnerships sometimes generate benefits, while also documenting negative instances. A partnership should distinguish conflict about scientific evidence, conflict about community interests, and conflict about distribution of resources. Procedures for recording minority views and revising a protocol provide a more reliable indicator of epistemic openness than the mere absence of disputes.
Partnership Synergy and the CBPR Conceptual Model
Partnership synergy describes the added capacity produced when diverse knowledge, skills, connections, and decision perspectives are combined. In Lasker et al. (2001), the proposition is that collaboration can produce an advantage unavailable to isolated actors, provided that its distinctive mechanism is defined and measured. Synergy must not be inferred merely from a partnership's existence or the number of participating organizations. Stronger evidence compares the range and quality of alternatives considered, the generation of feasible solutions, the relevance of research measures, and the capacity to mobilize resources across settings. It also asks whether coordination costs, conflict, or unequal influence offset those benefits. The conceptual claim has value precisely because it yields conditional empirical predictions.
The widely used CBPR conceptual model specifies four interrelated domains: context, partnership dynamics, research and intervention, and outcomes. Context includes political and historical conditions, community and university capacity, prior relations, and the problem being investigated. Partnership dynamics include structure and relationships, governance and resources, trust, and mutual learning. The research domain describes collaborative design and implementation. Outcomes range from changes in knowledge and capacity to policy, community conditions, and health. The model is neither an automatic causal chain nor an outcome guarantee; each proposed pathway requires independent measurement and attention to alternative explanations Belone et al. (2016) Oetzel et al. (2018).
Oetzel et al. (2018) empirically tested this model with data from federally funded US projects, including surveys from 200 principal investigators and 450 academic or community respondents, together with qualitative case studies. Structural-equation analyses supported associations linking context, partnership structures and dynamics, synergy and involvement, and reported intermediate and distal outcomes. The study provides unusually direct evidence for the proposed organization of CBPR partnership processes. Its observational and self-report structure restricts causal interpretation; participation in federally funded projects may affect the range of settings represented. The study should be read as a test of model coherence and association, while prospective design, independent outcome measurement, and intervention-based contrasts remain necessary for stronger causal identification.
Participation as an Operational and Measurable Construct
Research reports commonly label a study participatory without detailing who made which decisions. A valid participation measure should distinguish access, influence, authority, continuity, and accountability. Access records attendance or opportunity to contribute; influence records whether contributions change plans; authority records formal decision rights; continuity shows participation across phases; accountability measures how representatives report back to affected groups. Instruments can combine partner reports, project documents, observed meetings, financial records, and traces of decision revisions. Triangulation matters because researchers and community members may rate the same collaboration differently. An average partnership score can hide concentrated authority in one organization or the exclusion of particular populations.
Psychometric evaluation demands an explicit model of the construct, item development with community participation, response-process interviews, internal consistency where appropriate, test-retest reliability for stable attributes, and evidence of measurement invariance across partners and settings. Trust, decision power, and shared knowledge may change at different rates, so forcing them into one undifferentiated scale can be misleading. Dyadic or multilevel models should accommodate respondents nested within organizations and projects. Comparing projects requires attention to their historical and regulatory conditions. The empirical framework of Belone et al. (2016) illustrates why face validity must be tested with community partners; Israel et al. (2026) identifies the growth of validated evaluation tools as a development of the field.
Process fidelity is a separate question from intervention fidelity. A group might implement an evidence-based clinical intervention precisely while offering communities little influence over its design; another might demonstrate robust participatory governance while the intervention itself is poorly delivered. Outcome evaluation should measure both. Process logs may indicate whether translated materials were revised after community feedback, whether staffing rules changed after an advisory vote, or whether data interpretation meetings meaningfully altered a conclusion. Evidence of fidelity to CBPR principles strengthens descriptive claims about participation. It cannot alone show that participation improved mental health, reduced inequity, or caused policy reform: those outcomes require designs appropriate to their own inferential targets Yau et al. (2024) Cargo & Mercer (2008).
Community Definition, Sampling, and Recruitment
Sampling begins with an explicit account of the population about which a claim will be made. For geographic communities, the relevant frame may be residential addresses, neighborhoods, school districts, or service catchments. For identity-based and digitally networked communities, no administrative boundary may approximate membership. Community partners can identify hidden sampling failures, but partner networks themselves are often selective. Researchers should map eligibility rules, recruitment channels, differences between participants and nonparticipants, and the potential influence of gatekeepers. Recruitment success, representativeness, and community legitimacy are distinct measurements. A large sample obtained through one trusted organization may be valuable for particular aims while failing to support inferences about residents who distrust that organization Israel et al. (1998) Cargo & Mercer (2008).
Participation can change response behavior as well as participation rates. Local interviewers may establish familiarity and improve comprehension, but participants may feel uncomfortable disclosing criticism of local institutions to a recognizable neighbor. Translation can increase access while altering the connotations of psychological terms. Compensation may facilitate equitable inclusion or create undue pressure where resources are scarce. These effects should be investigated through cognitive interviews, pilot studies, comparison of recruitment streams, response-rate accounting, sensitivity analyses, and independent audit of field notes. Explicit documentation permits replication of procedures in a new context even when local adaptations differ.
Sample size calculations and qualitative adequacy address different inferential purposes. For a randomized or comparative quantitative design, prospective precision, power, clustering, and anticipated attrition matter. For qualitative interviews, the goal concerns depth, diversity of perspectives, thematic complexity, and transparent justification of what was sampled. Mixed-method projects need a plan explaining how sampling and timing connect the components. Partnerships should negotiate feasible recruitment while preserving the reasoning that makes the resulting evidence interpretable. The central ethical requirement is that a community should understand the practical consequences of the design's limits before strong claims about that community are published.
Collaborative Qualitative Inquiry and Participatory Observation
Qualitative CBPR often uses interviews, focus groups, participant observation, community mapping, diaries, oral histories, or collaborative interpretation workshops. Community researchers may adapt interview language, identify meaningful situations, or challenge outsider readings of observed events. Such involvement strengthens interpretive relevance when the analytic record captures how interpretations changed, which evidence supported them, and where disagreement persisted. Reflexive field notes should document the researcher's position, relationships, and practical influence on what people are willing to say. Comparing accounts across participants and organizations can test the stability of a theme without assuming that any one person speaks for a whole community.
Photovoice, described by Wang & Burris (1997), offers a particularly clear example of community knowledge production. Participants document local strengths and concerns photographically, discuss their images collectively, and may present findings to decision makers. Photographic evidence can make environmental barriers and everyday practices legible in ways that standardized survey categories miss. Interpretation still requires attention to whose images were selected, whose narration accompanies them, which people consent to being photographed, and whether public display exposes sensitive locations or relationships. Images do not interpret themselves; the analytic product is a negotiated, auditable account of their context and meaning.
Collaborative coding can combine inductive interpretation with explicit reliability procedures. Teams can first establish an analytic question, develop a provisional codebook, code material independently when suitable, then compare disagreement and document revisions. The aim of this exercise is to produce a traceable argument grounded in material rather than a superficial numerical consensus. Community researchers may also propose categories that academic team members initially overlook; these are analytic hypotheses to be checked against the full material. Presenting divergent interpretations, negative cases, and explanatory conditions helps a final report remain credible to both community and scholarly audiences. Macaulay et al. (1999) and Israel et al. (1998) describe the broader partnership rationales informing such cooperation.
Experimental and Quasi-Experimental Designs in Participatory Settings
CBPR can coexist with randomized trials, natural experiments, interrupted time series, matched comparisons, and longitudinal cohort designs. The participatory dimension concerns the production and governance of inquiry; causal identification concerns how a design distinguishes a treatment effect from other processes. Randomization, when ethically and operationally appropriate, can separate intervention assignment from baseline confounders, while community deliberation may shape the intervention, relevant endpoints, recruitment, and implementation conditions. At the cluster level, randomization must account for small numbers of clusters and correlation within schools, neighborhoods, or agencies. Community involvement should strengthen rather than obscure the rationale for estimators, uncertainty intervals, and predefined outcome definitions Wallerstein & Duran (2010) Cargo & Mercer (2008).
In many settings random allocation is infeasible, politically contested, or incompatible with service obligations. Then researchers may use matched sites, staggered rollout, comparative interrupted time series, synthetic controls, or other quasi-experimental strategies, provided that assumptions and diagnostics are reported. Community partners can identify historical events, migration, policy changes, or informal spillovers that threaten identification. Network relationships often violate the assumption that one unit's treatment has no influence on another. An intervention that changes organizational practice may affect comparison residents as well as nominally treated participants. Evaluators should specify the exposure, the relevant units, the expected spillover pathways, and sensitivity tests rather than silently applying individual-level independence assumptions.
A trial of a CBPR-developed intervention answers a question about that intervention, not automatically about the added causal contribution of participation. Demonstrating the incremental effect of CBPR would require a credible comparison between more and less participatory development pathways, or designs capable of isolating participatory mechanisms. Such comparisons encounter ethical, feasibility, and standardization difficulties: participation is relational and context-sensitive, and one partnership cannot simply be assigned a standardized dose. A useful alternative is to combine outcome evaluation with process and mechanistic analysis, test predicted mediators prospectively, and report uncertainty about pathways. The distinction prevents studies of successful interventions from being misread as controlled demonstrations that participatory governance itself caused success Oetzel et al. (2018) Jagosh et al. (2012).
Mixed Methods, Triangulation, and Process Evaluation
Mixed-method CBPR uses different forms of evidence to answer connected questions. Administrative records can identify who received services, surveys can measure symptoms or attitudes, observational data can document implementation, and interviews can explain why some groups refused to participate. Joint displays can show where qualitative mechanisms agree with or challenge quantitative patterns. Integration requires a stated design: a qualitative phase may inform measurement, parallel phases may examine divergent interpretations, or follow-up interviews may investigate unexplained quantitative results. Simply including both surveys and interviews is insufficient when their findings are never systematically related. Cargo & Mercer (2008) emphasizes an integrative approach to evaluating partnership and research outcomes.
Process evaluation should observe what the partnership actually does: recruitment and turnover, procedural fairness, meetings and participation, conflicts, protocol amendments, training, compensation, and dissemination actions. Early outcomes can include shared skills, improved research fit, and durable cross-organizational relationships. Later outcomes may involve services, policy, population health, or the distribution of environmental resources. Timing is consequential because a partnership can produce new organizational capacity long before health endpoints shift, and health outcomes may reflect unrelated reforms. A theory of change should identify expected lags and testable intermediate outcomes. A successful report states which elements are directly observed, which are self-reported, and which are inferred.
Independent perspectives strengthen evaluation. Academic investigators may overestimate community influence; paid partners may fear reporting dissatisfaction; community organizations may favor visible local outputs over standardized outcome measures. Confidential assessments, independent facilitation, anonymized partner surveys, external document review, and transparent qualitative negative-case analysis can reveal these discrepancies. Jagosh et al. (2012) offers a realist approach by investigating context–mechanism–outcome patterns rather than treating all partnerships as interchangeable. Realist explanation does not substitute for causal identification, but it can clarify why an intervention or partnership generates value in one configuration and disappointment in another. The strongest program of research links mechanistic propositions to designs capable of challenging them.
Foundational Empirical Cases and Community-Generated Evidence
Early case work remains central to the methodological development of participatory inquiry. Schulz et al. (1998) described a participatory survey on Detroit's east side for a community health intervention, illustrating the design challenges of collaborating on data collection in a specific social setting. Macaulay et al. (1999) drew on participatory research in primary care and Indigenous communities to show how affected groups can shape the production and uses of knowledge. These publications provide direct evidence that partnership procedures can be organized, documented, and critically examined in actual projects. They do not constitute randomized evaluations of the comparative health effect of CBPR. Their evidentiary contribution concerns feasible process, contextual knowledge, practical problem definition, and the ethical complexity of joint research.
Photovoice provides another research lineage in which the method itself creates opportunities for community interpretation. Wang & Burris (1997) described photographic elicitation and community discussion as tools for needs assessment and communication with policymakers. Its significance for psychology extends beyond visual documentation: it is a procedure for eliciting perceptions of place, symbolic meaning, perceived control, and shared understandings of local conditions. Investigators can examine which images and narratives recur, how interpretations change during discussion, and how presentations affect decision makers. Such designs must still test whether participants were selected fairly and whether changes attributed to the method follow from the documented sequence of activities.
Community-generated evidence also informs intervention design. Israel et al. (2010) described capacity-building and policy advocacy focused on health disparities, while Wallerstein & Duran (2010) argued that participatory processes can improve the practical translation of intervention research. These accounts identify plausible bridges between locally grounded explanations and program delivery. A critical analysis must distinguish the existence of a policy advocacy activity from evidence of durable policy change and from evidence that the policy caused improvements in population outcomes. Across case examples, process documentation often exceeds the availability of strong counterfactual outcome evaluation. The appropriate synthesis preserves those different evidentiary strengths rather than forcing a single quality label onto complex projects.
Systematic Reviews and the Evidence for Effects
The formal research literature includes evidence syntheses covering different populations, interventions, and definitions of engagement. The 2004 assessment by Viswanathan et al. (2004) helped establish the need to evaluate outcomes and methodological challenges systematically. Jagosh et al. (2012) undertook a realist review: from thousands of screened records, its final synthesis examined twenty-three participatory partnerships described across 276 publications. It identified recurrent mechanisms of appropriateness, recruitment, capacity development, conflict negotiation, sustainability, and system change, alongside adverse and unsuccessful pathways. These are explanatory patterns within a purposively assembled body of evidence; they should not be converted into an unconditional population-average treatment effect.
O’Mara-Eves et al. (2015) conducted a meta-analysis of community engagement in public-health interventions for disadvantaged groups. The review is relevant because engagement is one potential feature of CBPR interventions, but its inclusion rules encompass a broader family of practices. The resulting estimates therefore cannot be assigned without qualification to CBPR as defined here. Cyril et al. (2015) reviewed 24 studies of community engagement in disadvantaged populations and found beneficial results in 21, while reporting variable quality and levels of involvement. This finding describes the review's selected studies, with heterogeneous designs and outcomes; it does not mean that 87.5 percent of all CBPR projects are effective. Conclusions must consider selection, publication, measurement, and confounding before attributing improvements to participatory processes.
The systematic review by Yau et al. (2024) addressed CBPR interventions and mental-health outcomes more directly. It searched five databases for studies published from January 2000 through December 2022 and included 14 studies under its eligibility rules. Its four-dimensional framework examined community involvement in research, shared decision-making, capacity building, and attention to the original community health problem. The review sharpens a crucial methodological issue: the intensity of CBPR participation can be measured alongside intervention effects rather than inferred from a label. The comparatively small selected literature, diverse programs, and variation in fidelity make universal clinical claims inappropriate. These findings support a research agenda for stronger reporting, comparative designs, and longitudinal mental-health assessment.
Models Tested with Partnership Data
The study by Oetzel et al. (2018) is a central empirical test because it places partnership organization, rather than an individual clinical intervention, at the center of analysis. Investigators identified 294 federally funded US CBPR projects, obtained project questionnaires from 200 principal investigators, and received survey data from 450 academic or community participants and principal investigators. Structural-equation modeling supported the expected positive relations among context, structures and dynamics, partnership synergy and community participation, and subsequent reported outcomes. Qualitative case studies added detail on pathways. The result provides convergent evidence for a conceptual map of partnership functioning across many projects rather than a single success story.
Several design constraints matter. Participation in a federally funded project introduces selection that can privilege institutions with grant-writing capacity and stable partnerships. Reports about trust, capacity, and perceived success may share informant and method variance. Cross-sectional associations may reflect reverse causality: a successful project could itself improve reported trust and synergy. Structural equation modeling does not eliminate omitted confounding or prove that measured arrows follow the causal direction depicted in a conceptual framework. Seven qualitative cases can clarify the processes hypothesized, but the existence of a vivid case does not make the underlying comparative effect experimentally identified Oetzel et al. (2018).
The earlier consultation study by Belone et al. (2016) evaluated the model with focus groups from six partnerships. Community respondents affirmed and modified important constructs, especially trust development, mutual learning, capacity, and power. The evidentiary contribution was face validity and conceptual refinement rather than prediction of population health. Together, these investigations establish a research progression: communities help identify meaningful constructs, researchers operationalize them, and multisite data test relations among the resulting measures. The next stage requires longitudinal data, independent outcomes, transparent model comparisons, and careful attention to the extent that findings transfer across policy systems and community histories.
Health Equity, Mental Health, and Institutional Change
Health equity is often the practical horizon of CBPR because inequalities are shaped by housing, exposure, discrimination, service organization, income, and institutional trust. Participatory inquiry may reveal mechanisms missing from individual-level intervention models and produce culturally meaningful outcomes. Wallerstein & Duran (2010) situates the approach within translation and health-equity research; Israel et al. (2010) details how coalition capacity can contribute to policy advocacy. The pathway from local knowledge to population-level benefit nevertheless has several stages: problem recognition, feasible intervention design, implementation, changes in practice or policy, and sustained distribution of benefits. Evidence at one stage cannot be silently extended to the others.
Mental-health applications include the contextual interpretation of distress, social support, service access, stigma, and culturally adapted prevention. A collaborative research process can help specify what participants recognize as meaningful changes in function and well-being. Yet validated symptom measures, diagnostic consistency, adverse outcomes, and comparison conditions remain essential where clinical effectiveness is at issue. Yau et al. (2024) provides a focused evidence synthesis, while the broader reviews O’Mara-Eves et al. (2015) and Cyril et al. (2015) require careful separation of community engagement from the specific structure of CBPR. The research tradition offers promising routes for intervention development; the magnitude and durability of health benefit remain population- and intervention-dependent.
Institutional change introduces a different endpoint from symptom reduction. A community partnership may establish a lasting advisory body, redistribute funds, change eligibility rules, or strengthen data governance without producing measurable short-term changes in a health scale. These outcomes deserve their own indicators and follow-up periods. Conversely, an improved service metric can occur under a partnership that remains unequal. Evaluation should record what changed, for whom, by whose decision, for how long, and through which documented mechanism. The 2026 historical synthesis Israel et al. (2026) identifies policy and structural transformation as major directions in the field. A rigorous account translates that agenda into measurable organizational and population outcomes without treating declared commitment as proof of achievement.
Community Knowledge, Cultural Translation, and Indigenous Partnerships
The idea of community expertise becomes scientifically consequential only when its contribution to design and inference can be identified. Community researchers can recognize local categories, relationships, institutional histories, and environmental constraints that standardized investigators misclassify. Yet local knowledge, like professional expertise, has limits: a personally salient experience may be rare, an organizational leader may interpret collective interests selectively, or a familiar explanation may conflict with measured evidence. Productive partnership therefore establishes procedures for bringing diverse forms of knowledge into contact. Community members can propose hypotheses, identify hidden assumptions, and examine interpretations; researchers can communicate alternative causal models, measurement limitations, and the consequences of selective sampling. Validity arises from the critical relation among claims, observations, and tests, rather than the social identity of the person offering an explanation Israel et al. (1998) Macaulay et al. (1999).
Cross-cultural research particularly benefits from examining how categories travel. Words for distress, agency, belonging, safety, and treatment expectations may organize experience differently across languages and social institutions. A nominally accurate translation may fail to preserve an item’s meaning, while a locally appropriate expression may change the construct being measured. Community collaboration can supply cognitive interviewing, conceptual mapping, alternative item phrasing, and feedback on the social consequences of questionnaires. Researchers should then test response distributions, factor structure, differential item functioning, and criteria validity where sample size permits. The strongest result documents both the reasons for adaptation and evidence that the adapted instrument measures its intended object. Belone et al. (2016) illustrates community participation in construct refinement; Oetzel et al. (2018) indicates how partnership measures may subsequently enter larger empirical analyses.
Indigenous research contexts raise further questions about historical extraction, tribal or community jurisdiction, collective knowledge, and obligations extending beyond an individual research consent form. The precise arrangements must be established with the relevant Indigenous partners and applicable governance protocols, rather than inferred from a universal template. A project should identify the legitimacy of participating institutions, distinguish collective data stewardship from individual privacy rights, specify secondary-use permissions, and record how results can be challenged or corrected. The comparative research question is whether these arrangements improve interpretive accuracy, fair allocation of benefits, long-term access to evidence, and the community’s capacity to conduct research on its own terms. Examples of participatory inquiry involving Indigenous communities already formed part of the historical literature surveyed by Macaulay et al. (1999).
Contextual Heterogeneity and the Transportability of Evidence
A CBPR partnership is constituted by a local combination of resources, institutional authority, social trust, and scientific capacities. The same meeting rules or advisory structure can function differently under different histories of conflict, political representation, professional relationships, or public-service organization. Consequently, transferring a successful procedural model from one setting to another requires a hypothesis about which contextual properties activate its mechanisms. Analyses should specify site-level covariates, the composition of participating organizations, policy constraints, baseline conditions, and the degree of variation within a site. The context–mechanism–outcome approach in Jagosh et al. (2012) provides a useful explanatory vocabulary for this problem, and the conceptual model examined in Oetzel et al. (2018) supplies operational constructs that can be tested in new settings.
Generalizability involves several questions. Would a particular estimate apply to people who did not join the study? Would an intervention effect recur in a different community under similar implementation conditions? Would participatory governance yield the same research quality if a different coalition held authority? These are distinct inferences, even when a report describes all of them as scalability. A program can be locally successful because trusted individuals provide an unusually effective bridge across institutions; replicating a written protocol without those relationships may not reproduce the mechanism. Rather than treating the original context as noise, researchers can report the conditions that appear necessary and compare outcomes after deliberate adaptation. Cross-site studies should measure both common constructs and locally specific processes.
Transportability is particularly difficult when effects are mediated by networks. Changing a service provider's policy may influence people who never participated, while a local information campaign may diffuse into comparison sites through social contacts. Such interference can be a desired outcome for communities and an identification problem for evaluators. Research protocols need to define exposure, units, plausible spillover channels, and the target estimand. Repeated measurements of organizations and communities can distinguish immediate research engagement from delayed institutional consequences. The 2026 field synthesis Israel et al. (2026) emphasizes policy and structural change; rigorous evaluation requires designs that can follow these pathways without exaggerating their causal certainty.
Failures, Unintended Consequences, and Publication Bias
Research programs that advocate a socially valued method have a strong incentive to emphasize exemplary success. Yet a participatory partnership can intensify local disputes, empower established organizations over marginalized members, create false expectations of reform, or expose identifiable groups to stigma. Community researchers may feel pressure to endorse conclusions that secure continued funding. Academics may favor rapidly publishable outcomes over the time needed for genuine deliberation. Studies should actively seek negative cases, report changes in governance, and disclose when partnership arrangements were terminated or modified. The realist synthesis Jagosh et al. (2012) describes unwanted consequences and difficult negotiations as part of the empirical record rather than deviations to be excluded from analysis.
Publication bias can arise at several stages. Funders may select partnerships with strong administrative support; investigators may recruit accessible community organizations; projects with favorable results may receive more publication opportunities; and review searches may miss technical reports or work disseminated primarily in local languages. Quantitative effects can also be distorted by selective outcomes, multiple analyses, attrition, and the absence of independent replication. A research portfolio should prospectively specify major outcomes, retain documentation of unsuccessful implementation, and make protocols and permitted analytic code available where appropriate. Reviews such as Cyril et al. (2015) and O’Mara-Eves et al. (2015) must be read in light of heterogeneity and study quality. The quantity of positive reports alone is an inadequate estimate of the prevalence of genuine benefit.
Unintended consequences should be measured at more than one level. Individuals may report increased research burden even as an organization gains capacity; an intervention may improve service access for a majority while worsening it for people outside partner networks; a policy victory may shift costs toward another community. Transparent evaluation can assess burden, exclusion, opportunity cost, long-term trust, and who benefits from a given institutional change. Findings remain ethically and scientifically informative when they contradict the partnership's preferred theory of change. This commitment distinguishes a research partnership from a public-relations coalition and provides the conditions for cumulative learning across diverse communities Cargo & Mercer (2008) Israel et al. (2026).
Research Quality, Reproducibility, and Reporting Standards
Quality appraisal begins by separating four evaluative dimensions: participatory legitimacy, integrity of empirical measurement, validity of inference, and the consequences of research. A project may excel on one and fail on another. Its partnership could provide equitable representation but use a poorly validated outcome scale; it could employ a strong randomized intervention design without meaningful shared authority; it could produce valuable public data while allocating its burdens unfairly. A comprehensive report should independently describe each dimension so that the judgment of scientific credibility does not become a proxy for the moral appeal of the project. Oetzel et al. (2018) and Belone et al. (2016) show the feasibility of measuring partnership constructs; Yau et al. (2024) indicates the value of separately scoring adherence to participatory principles when reviewing intervention outcomes.
Reproducibility does not mean exporting an unchanged intervention into every community. It means making the evidence-generating process sufficiently inspectable for another team to understand the decisions, constraints, transformations, and analyses that produced a claim. A CBPR audit trail can preserve protocol versions, community amendments, recruitment records, sampling frames, interview guides, codebooks, quantitative analysis plans, consent changes, and a record of which interpretations were disputed. Where privacy or collective governance prevents public release of raw data, investigators can document access conditions and share de-identified aggregates, synthetic test material, and code when appropriate. A community-approved explanation of restricted access may itself be part of transparent research stewardship.
Reporting standards should be chosen according to the actual study design. A randomized trial requires the reporting relevant to its assignment, exclusions, harms, and statistical analysis; a qualitative inquiry requires a transparent account of selection, researcher positioning, data generation, interpretation, and negative cases; a systematic review requires a reproducible search and synthesis protocol. Because this monograph uses a critical narrative synthesis, it makes no claim of PRISMA compliance, exhaustive database search, pooled effect calculation, or original participant data. Instead, it grounds each synthesis in identifiable publications and marks inferential boundaries. This principle becomes even more important as digital research platforms and Artificial systems make it easier to generate apparently comprehensive summaries without transparent research histories Cargo & Mercer (2008) Israel et al. (2026).
Ethics, Collective Consent, and Data Stewardship
Participatory ethics extends the individual informed-consent encounter into the continuing relationships through which research questions, risks, and outputs are negotiated. Individual consent remains indispensable when personal information or interventions are involved, but the effects of research may also fall on identifiable groups, local institutions, and people who never enrolled. Publishing a map of stigmatized behavior, for example, may identify a small community despite the removal of names. Ethical review should therefore examine collective consequences, representation, appropriate data access, withdrawal procedures, and the governance of secondary use. Community consultation does not automatically authorize access to every individual's information, and institutional ethics approval does not alone establish community legitimacy Macaulay et al. (1999) Israel et al. (2026).
Data stewardship is a practical allocation of rights and responsibilities. Partnership agreements should specify custody, storage jurisdiction, access logs, retention, secondary analyses, de-identification limits, authorship, and procedures for contesting public interpretation. Where Indigenous peoples are research partners, established collective rights, tribal protocols, and community-specific data-governance principles require careful attention. A generic claim that all data are jointly owned obscures the differences among individual privacy, institutional legal responsibilities, and collective stewardship. Open science is strengthened when permitted data and analytic code can be shared with clear provenance; privacy and community obligations may require controlled access, synthetic examples, or transparent reasons for restricted release.
Ethical evaluation includes the distribution of research burdens. Community partners may absorb reputational risk, emotional labor, travel time, or conflict within their networks. Compensation and credited intellectual contribution should be negotiated, while recruitment and consent materials should clearly distinguish research from clinical services or promised policy changes. The right to criticize project findings is central to accountability, including when conclusions challenge the partnership's preferred narrative. Disputes about validity should be resolved with evidence and documented analysis, while disputes about public identification require specific harm assessment. These provisions make participatory ethics an ongoing institutional practice that can be audited rather than a single approval event Cargo & Mercer (2008) Israel et al. (1998).
Institutional Sustainability, Policy Translation, and Economic Constraints
Sustaining CBPR involves continuity of personnel, resources, authority, and useful relationships after initial funding. A successful short grant may produce an excellent report and leave a community without access to the infrastructure needed to act on it. Indicators of sustainability include independent community leadership, durable access to research tools, maintenance of relationships during gaps in funding, policy uptake, and subsequent projects initiated by local partners. The realist synthesis of Jagosh et al. (2012) identified continuity and unanticipated system change among plausible partnership benefits, while showing that context matters. A credible sustainability claim therefore requires follow-up beyond the funded intervention and attention to which benefits remain under community control.
Policy translation is a causal sequence across different institutions. An evidence brief can influence deliberation, but administrative adoption depends on budgets, legal authority, political incentives, staff competence, and public contestation. Researchers should document meetings, decisions, legislative or organizational changes, and implementation milestones instead of treating the number of public presentations as an outcome equivalent to policy change. In multi-sector coalitions, distributed ownership may aid continuity but complicate accountability. The appropriate evaluation design follows the specific institutional mechanism at work and measures intended as well as unintended consequences Israel et al. (2010) Israel et al. (2026).
The economic costs of participation deserve explicit estimation. Consensus-building, local translation, training, shared analysis, and conflict mediation require time and money. These expenditures may improve research relevance and later uptake, but the counterfactual must be made explicit: what would an otherwise comparable research process cost and achieve? A full evaluation can record direct funding, community time, opportunity costs, delays, and the distribution of future benefits. Such accounting resists the temptation to celebrate participatory work as inherently costless or efficient. In a responsible program, partnership investments are justified by the scientific and social outcomes they help produce, and uncertainty about their magnitude is reported.
Digital Participation and Research in the Artificial Era
Digital tools transform how communities form, deliberate, and participate in research. Remote focus groups, collaborative maps, online diaries, accessible survey platforms, and shared analytic workspaces can widen the geographic reach of inquiry. At the same time, connectivity, platform literacy, disability access, language interfaces, and algorithmically curated visibility alter who can take part. A digital community may have legitimate members without a common territory, while a neighborhood research project may exclude residents who do not possess reliable devices or private space for online participation. Digital CBPR should therefore audit selection into platforms and document how technological mediation affects consent, discussion, and representativeness.
Artificial systems may assist with translation, coding, transcript organization, accessibility, or the comparison of competing interpretations. Their use raises operational questions about error, privacy, traceability, differential performance across languages, and the delegation of interpretive authority. A machine-generated thematic summary is a proposed representation of recorded material that requires human and community evaluation. Automated analytic procedures should log model versions, prompts or methods where appropriate, data-processing restrictions, and error audits. Community partners should have an intelligible opportunity to challenge conclusions produced through computational mediation. These procedures preserve the evidentiary standards of CBPR while recognizing the technological conditions of the Artificial Era.
Future human–Artificial research partnerships may introduce new forms of collective coordination, memory, and decision support. These are prospective questions requiring their own research designs rather than presently established CBPR outcomes. The theoretical questions include who authorizes an artificial participant to act, whether its recommendations are independently contestable, how accountability persists after system updates, and how organizational memory survives platform change. Operational comparisons could test whether AI assistance changes inclusion, meeting deliberation, translation quality, and partner trust. The decisive research criterion remains the empirical relationship between participation, knowledge production, and consequences for the affected community.
Critical Scientific Discussion
The central empirical conclusion is conditional. CBPR provides a coherent organizational framework for combining experiential, professional, and scientific knowledge, and its partnership processes can be measured and examined. Conceptual and qualitative investigations identify trust, mutual learning, power, representation, and synergy as meaningful constructs Belone et al. (2016) Lasker et al. (2001). Multisite observational evidence supports associations among partnership structure, dynamics, engagement, and reported outcomes Oetzel et al. (2018). Reviews describe benefits that depend on context while documenting methodological heterogeneity and unsuccessful partnerships Jagosh et al. (2012) Cyril et al. (2015). Evidence for the causal contribution of participation to particular population-health outcomes is more constrained than evidence that participation changes research processes.
Five unresolved questions should direct subsequent investigations. Researchers need stronger longitudinal and comparative identification of partnership mechanisms; participation measures with demonstrated cross-cultural and organizational validity; sampling designs that capture dissent and underrepresented groups; methods for measuring structural and policy outcomes over long horizons; and procedures for preserving community data stewardship while maintaining transparent scientific reporting. Multilevel models, process tracing, pre-registered intervention evaluations, community-reviewed instruments, and independent replication can address parts of these problems. New projects should report negative findings and implementation failures as carefully as successes, because selection on published achievements makes theories of participatory benefit appear more universal than the evidence warrants.
Conclusion
Community-Based Participatory Research is consequently a scientific program about the governance and production of knowledge under conditions of social interdependence. It treats communities as sources of substantive expertise and legitimate participants in research decisions, and it places the resulting inquiry under evidence-sensitive scrutiny. Its long-term significance lies in linking research validity with contextual relevance, reciprocal capacity, and accountability to those whose lives are investigated. The historical trajectory from foundational partnership principles to modern conceptual models and outcome reviews demonstrates both achievement and a remaining program of rigorous tests Israel et al. (1998) Israel et al. (2026) Yau et al. (2024). The finished synthesis locates this subject within Community Psychology and its methodological, cultural, and health-related connections without dissolving its distinctive research object.
The practical criterion that closes this investigation is correspondence between the authority a partnership publicly claims and the decisions, evidence, and consequences that can be independently traced. Future CBPR should publish a transparent account of inclusion, power, analytic validity, contextual adaptation, negative outcomes, and community use of findings. The quality of a participatory study is established by how these commitments are performed under scrutiny and by what its methods can genuinely support.
References
- Belone, L., Lucero, J. E., Duran, B., Tafoya, G., Baker, E. A., Chan, D., Chang, C., Greene-Moton, E., Kelley, M., & Wallerstein, N. (2016). Community-based participatory research conceptual model: Community partner consultation and face validity. Qualitative Health Research, 26(1), 117–135. https://doi.org/10.1177/1049732314557084
- 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
- Collins, S. E., Clifasefi, S. L., Stanton, J., The LEAP Advisory Board, Straits, K. J. E., Gil-Kashiwabara, E., Rodriguez Espinosa, P., Nicasio, A. V., Andrasik, M. P., Hawes, S. M., Miller, K. A., Nelson, L. A., Orfaly, V. E., Duran, B. M., & Wallerstein, N. (2018). Community-based participatory research (CBPR): Towards equitable involvement of community in psychology research. American Psychologist, 73(7), 884–898. https://doi.org/10.1037/amp0000167
- Cyril, S., Smith, B. J., Possamai-Inesedy, A., & Renzaho, A. M. N. (2015). Exploring the role of community engagement in improving the health of disadvantaged populations: A systematic review. Global Health Action, 8, 29842. https://doi.org/10.3402/gha.v8.29842
- 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
- Israel, B. A., Coombe, C. M., Cheezum, R. R., Schulz, A. J., McGranaghan, R. J., Lichtenstein, R., Reyes, A. G., Clement, J., & Burris, A. (2010). Community-based participatory research: A capacity-building approach for policy advocacy aimed at eliminating health disparities. American Journal of Public Health, 100(11), 2094–2102. https://doi.org/10.2105/AJPH.2009.170506
- 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
- 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. The Milbank Quarterly, 90(2), 311–346. https://doi.org/10.1111/j.1468-0009.2012.00665.x
- Lasker, R. D., Weiss, E. S., & Miller, R. (2001). Partnership synergy: A practical framework for studying and strengthening the collaborative advantage. The Milbank Quarterly, 79(2), 179–205. https://doi.org/10.1111/1468-0009.00203
- Macaulay, A. C., Commanda, L. E., Freeman, W. L., Gibson, N., McCabe, M. L., Robbins, C. M., & Twohig, P. L. (1999). Participatory research maximises community and lay involvement. BMJ, 319(7212), 774–778. https://doi.org/10.1136/bmj.319.7212.774
- O’Mara-Eves, A., Brunton, G., Oliver, S., Kavanagh, J., Jamal, F., & Thomas, J. (2015). The effectiveness of community engagement in public health interventions for disadvantaged groups: A meta-analysis. BMC Public Health, 15, 129. https://doi.org/10.1186/s12889-015-1352-y
- 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, 7281405. https://doi.org/10.1155/2018/7281405
- Schulz, A. J., Parker, E. A., Israel, B. A., Becker, A. B., Maciak, B. J., & Hollis, R. (1998). Conducting a participatory community-based survey for a community health intervention on Detroit's east side. Journal of Public Health Management and Practice, 4(2), 10–24. https://doi.org/10.1097/00124784-199803000-00004
- Viswanathan, M., Ammerman, A., Eng, E., Garlehner, G., Lohr, K. N., Griffith, D., Rhodes, S., Samuel-Hodge, C., Maty, S., Lux, L., Webb, L., Sutton, S. F., Swinson, T., Jackman, A., & Whitener, L. (2004). Community-based participatory research: Assessing the evidence. Evidence Report/Technology Assessment (Summary), 99, 1–8. https://pubmed.ncbi.nlm.nih.gov/15460504/
- 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
- Wang, C., & Burris, M. A. (1997). Photovoice: Concept, methodology, and use for participatory needs assessment. Health Education & Behavior, 24(3), 369–387. https://doi.org/10.1177/109019819702400309
- Yau, J. H.-Y., Wong, E. L.-Y., Kanagawa, H. S., Liu, T., Wong, G. H.-Y., & Lum, T. Y.-S. (2024). Effectiveness of community-based participatory research (CBPR) interventions on mental health outcomes: A systematic review. Social Science & Medicine, 363, 117491. https://doi.org/10.1016/j.socscimed.2024.117491