03 · What You Need to Know
Combining Participatory Governance With Quantitative Research Methods
Why Is Participatory Research Sometimes Associated With Qualitative Methods?
Participatory research often involves dialogue, reflection, contextual interpretation, and collaboration with people affected by the research. Interviews, observations, workshops, and focus groups can support these activities.
Consequently, participatory approaches are frequently associated with qualitative inquiry.
However, participation is not itself a qualitative data collection technique. It concerns the relationship between researchers and those who contribute to defining, conducting, interpreting, or using the investigation.
A community may help develop a survey, determine which outcomes matter, or participate in the governance of a randomized trial.
The distinction is important because a study can be participatory in its governance while using quantitative evidence to answer its substantive question.
What Is the Difference Between Participatory Approach and Research Design?
Participatory approach
Concerns who participates in shaping research decisions, how authority is distributed, and how relevant knowledge and interests influence the study.
Research design
Concerns the structure of the investigation, including sampling, allocation, measurement, timing, comparisons, and the logic supporting the intended inference.
These dimensions can be combined in different ways.
For example, a community-engaged randomized trial may use shared decisions to select outcomes while preserving a valid random allocation procedure.
A participatory survey may involve community members in developing questions and interpreting patterns while using probability sampling and statistical estimation.
Participation does not determine whether a study is experimental, observational, qualitative, or mixed methods.
Which Quantitative Designs Can Be Participatory?
Several quantitative designs can incorporate meaningful stakeholder involvement, although the appropriate role depends on the study.
| Research design |
Possible participatory contribution |
Methodological requirement |
| Randomized controlled trial |
Co-design recruitment, outcomes, intervention delivery, and dissemination. |
Preserve valid allocation, comparisons, and analysis. |
| Cluster randomized trial |
Involve institutions or communities in intervention and implementation planning. |
Account for cluster allocation, correlation, and recruitment bias. |
| Cross-sectional survey |
Develop relevant questions and improve population access. |
Maintain measurement validity and appropriate sampling inference. |
| Longitudinal observational study |
Identify meaningful follow-up outcomes and acceptable contact procedures. |
Address temporality, missing data, and confounding. |
| Quasi-experimental evaluation |
Help identify feasible comparison conditions and implementation contexts. |
Evaluate the assumptions supporting causal identification. |
| Mixed-methods study |
Participate in developing and integrating qualitative and quantitative components. |
Specify the logic and procedures for integrating evidence. |
The participatory component should influence consequential decisions rather than simply provide a label attached to an otherwise conventional design.
Can a Randomized Controlled Trial Be Community-Based Participatory Research?
Yes. A randomized trial can be conducted within a community-based participatory research partnership when community members and researchers collaborate meaningfully across relevant stages of the study.
The partnership may help define the research question, determine whether randomization is acceptable, identify important outcomes, and develop implementation procedures.
Community partners may also contribute to interpretation and dissemination.
However, the trial must retain the methodological features necessary to support its intended causal conclusions.
For example, partners should not alter treatment assignments after learning participants' characteristics in ways that undermine the planned randomization.
Participatory governance and valid randomization can coexist when their responsibilities and boundaries are established prospectively.
How Can Stakeholders Influence a Trial Before Randomization?
Many consequential decisions occur before participants or clusters are allocated.
Stakeholders can help determine whether the intervention addresses an important problem, whether the comparison condition is acceptable, and whether the proposed outcomes capture meaningful consequences.
They can review recruitment procedures, identify participant burdens, and assess whether intervention activities can be delivered under realistic conditions.
These contributions may improve relevance and feasibility without changing the allocation mechanism.
However, recommendations affecting eligibility, sample size, measurement, or the intervention itself require methodological evaluation.
Stakeholder involvement should be incorporated into protocol development rather than introduced only after the essential trial decisions have been finalized.
What If Community Partners Oppose Randomization?
Opposition may arise from concerns about fairness, unequal access, trust, or uncertainty about the intervention's benefits.
Researchers should investigate the underlying concern rather than assume that community partners misunderstand scientific methods.
Depending on the research question and ethical circumstances, alternatives may include delayed intervention arrangements, cluster randomization, stepped-wedge designs, or nonrandomized evaluations.
However, these designs involve different assumptions and limitations. A stepped-wedge design, for example, introduces time-related considerations and is not automatically preferable or more ethical.
Where no acceptable randomized design is feasible, investigators may select another defensible design and appropriately limit causal claims.
Participatory involvement does not require researchers to present a nonrandomized comparison as equivalent to random assignment.
Can Community Partners Help Determine Randomization Units?
They may contribute important contextual knowledge when researchers decide whether to randomize individuals, classrooms, clinics, institutions, or communities.
For example, teachers may explain that students in the same classroom routinely share learning materials, making individual allocation vulnerable to contamination.
Cluster randomization might therefore be considered.
However, clustering affects sample-size requirements, statistical precision, recruitment, and analysis. Researchers must account for within-cluster correlation and the number of independent allocation units.
Community preference for group allocation is relevant but does not independently establish that a cluster randomized design is appropriate.
Can Stakeholders Help Select Quantitative Outcomes?
Yes. Stakeholders may identify outcomes that researchers overlook or undervalue.
For example, an educational intervention study may initially prioritize examination performance, while students identify accessibility, workload, or confidence as important additional consequences.
These perspectives can improve outcome relevance.
However, selecting outcomes also requires attention to construct validity, measurement quality, sample-size planning, and the intended estimand.
Stakeholders can participate in research outcome selection without every preferred outcome becoming a primary endpoint.
Important outcomes should ordinarily be prespecified before the results are known.
Can Participatory Research Use Standardized Questionnaires?
Yes. Standardized instruments can support consistent measurement within participatory studies.
Stakeholders may help evaluate whether questions are understandable, relevant, culturally appropriate, and accessible to the intended population.
However, modifying a validated questionnaire may alter its measurement properties.
Researchers should distinguish changes to administration that preserve the intended construct from substantive changes to item content or scoring.
When adaptation is necessary, appropriate validity evidence should be obtained.
Standardization and participation are compatible when the procedures preserve meaningful and defensible measurement.
Can Community Partners Help Design Sampling Procedures?
Community partners may identify populations that conventional recruitment methods overlook and explain why certain channels are inaccessible or distrusted.
They may also contribute to the design of sampling frames, contact procedures, and recruitment materials.
However, community networks do not automatically produce representative samples.
Recruiting exclusively through one organization may exclude eligible individuals who do not belong to that network.
Researchers should distinguish improving recruitment reach from establishing the sampling assumptions required for population inference.
Stakeholder involvement in participant recruitment methods should therefore be evaluated alongside selection bias and coverage.
Can Participatory Quantitative Research Still Estimate Causal Effects?
Potentially. Causal inference depends on the design, assumptions, and implementation of the study rather than whether stakeholders participate in its governance.
A suitably conducted randomized trial may support causal inference when allocation, follow-up, measurement, and analysis are appropriate.
Participatory observational research may also investigate causal questions, but it requires a defensible identification strategy addressing relevant confounding and other biases.
Stakeholder knowledge can improve causal reasoning by identifying institutional mechanisms, potential confounders, and relevant exposure versions.
However, stakeholder consensus about why an outcome occurred does not establish causation.
Causal conclusions must remain consistent with the evidence and assumptions of the study.
Can Stakeholders Help Specify Quantitative Analyses?
Yes. Stakeholders may help identify which comparisons are meaningful and which population differences deserve investigation.
For example, community partners may request separate estimates for groups that experience different levels of service access.
Researchers should evaluate whether the sample supports those analyses and whether the proposed comparisons are aligned with the research question.
Statistical decisions involving estimation, uncertainty, multiplicity, missing data, or causal identification may require specialized expertise.
Partners with appropriate expertise may participate directly in these technical decisions.
Shared decision-making should make analytical choices more transparent without allowing preferred results to determine the methods.
How Can Stakeholders Participate in Quantitative Data Interpretation?
Stakeholders may identify contextual factors that help explain observed patterns or unexpected results.
For example, a quantitative analysis may show lower intervention engagement at certain sites. Local partners may explain that these sites experienced connectivity problems or scheduling disruptions.
Such explanations can guide additional investigation and improve interpretation.
However, researchers should distinguish contextual hypotheses from mechanisms established by the data.
Post hoc explanations should not be presented as prespecified causal findings.
Stakeholder interpretation is most valuable when it complements, rather than replaces, appropriate statistical reasoning.
Does Participatory Research Require Mixed Methods?
No. Participatory research can use exclusively quantitative, exclusively qualitative, or mixed-methods designs.
A study becomes mixed methods when qualitative and quantitative components are intentionally integrated to address the research purpose, not merely because stakeholder meetings accompany a quantitative study.
For example, interviews conducted solely to plan recruitment do not necessarily constitute a formal qualitative research strand.
Researchers should describe stakeholder involvement and research methodology separately, explaining how they relate where appropriate.
Participation is a governance and knowledge-production arrangement, not an automatic classification of the study's data.
How Can Participatory Quantitative Studies Maintain Intervention Fidelity?
Stakeholders may recommend adaptations that improve acceptability or implementation.
However, uncontrolled changes to intervention components can create different treatment versions and complicate interpretation.
Researchers should identify which intervention functions must remain consistent and which operational features may be adapted.
For example, a digital learning intervention might permit locally relevant examples while preserving the intended feedback process and assessment requirements.
Permitted adaptations should be documented, and their consequences for fidelity and comparability should be evaluated.
Participatory design does not require identical implementation in every context, but variation should be scientifically understood.
How Do You Prevent Stakeholder Influence From Biasing Quantitative Findings?
Researchers should establish appropriate safeguards for allocation, measurement, data handling, analysis, and reporting.
These may include prespecified outcomes and analytical methods, standardized procedures where necessary, appropriate blinding, independent assessment, and transparent documentation of deviations.
Not every safeguard is feasible or required in every design.
For example, participants and implementers may be aware of allocation in a behavioral intervention, while outcome assessors can sometimes remain blinded.
Stakeholders should be able to challenge methodological decisions, but no partner should be permitted to manipulate data or suppress unfavorable results.
Scientific accountability should be established through the study's governance arrangements.
How Should Participatory Authority Be Allocated?
The partnership should identify which decisions are shared, which require consultation, and which involve formal scientific, ethical, or institutional responsibility.
Partners may jointly determine research priorities and acceptable recruitment procedures while designated investigators retain responsibility for protocol compliance and research integrity.
However, formal responsibility does not automatically justify excluding partners from meaningful methodological discussions.
The arrangement should reflect the research decisions partners can legitimately share and the procedures for maintaining scientific accountability.
Shared authority should be genuine, documented, and accompanied by appropriate support.
How Should Participatory Randomized or Quantitative Studies Be Reported?
Researchers should report both the scientific design and the participatory process.
For randomized trials, the applicable CONSORT reporting guidance addresses the trial's design, conduct, analysis, and results. For observational studies, STROBE may be relevant.
The GRIPP2 reporting checklists provide guidance for reporting patient and public involvement, including its aims, methods, and influence.
These frameworks serve different purposes and should not be treated as substitutes for one another.
Reports should identify what partners actually influenced, which decisions were shared, and how methodological requirements were maintained.
Researchers should avoid labeling a study participatory merely because stakeholders attended meetings or endorsed a completed protocol.
Watch Out
Do not confuse participatory governance with freedom to modify randomization, outcomes, or analyses after observing results. Meaningful stakeholder involvement can shape these decisions prospectively, but changes must remain scientifically justified and transparently documented.
