Can Researchers Recruit Their Own Clients or Service Users?

If Someone Depends on Your Service, Can They Freely Refuse Your Research?

Researchers do not work only in universities and hospitals.

Research may take place in counseling centers, social-service agencies, rehabilitation programs, charities, shelters, legal services, community organizations, employment programs, educational support services, government agencies, and many other settings where researchers or their institutions already provide something participants need.

This creates a recurring ethical question.

Can you ask your own clients or service users to participate in research?

Potentially. But the invitation may carry more weight than researchers realize when the person asking also controls, provides, recommends, or influences a valued service. A client may worry that saying no will affect eligibility, priority, quality of service, benefits, future assistance, or the professional relationship itself.

The central ethical task is therefore to keep the research decision separate from the person's access to services and ordinary treatment by the organization.

Clients Can Participate, but Services Must Not Become Leverage

Dependency Can Exist Wherever Someone Controls Something You Need

“Client” covers very different relationships

A client could be someone receiving psychological counseling, legal assistance, rehabilitation, social welfare support, employment services, disability services, housing assistance, financial counseling, community programs, or other professional support.

A service user might interact with a researcher once or depend on the organization for months or years.

These relationships are too diverse for a single rule based merely on the label “client.”

The useful question is what the researcher or organization controls.

Does the person depend on them for access to housing, therapy, food assistance, case management, legal help, educational support, benefits, program placement, documentation, or another important service?

The more consequential that dependency, the more carefully researchers should examine whether the power difference changes how freely the invitation can be refused.

Withholding an entitled service can be coercion

OHRP gives a clear general example of coercion in research: threatening loss of needed health services if someone refuses to participate. Its guidance also states that overt coercion can include threatening loss of services or access to programs to which potential participants are otherwise entitled.

The principle extends beyond healthcare.

If access to something the person is otherwise entitled to receive is threatened to obtain participation, the issue is not merely an overly persuasive invitation. It raises a coercion concern.

People may perceive consequences that nobody explicitly threatens

The more difficult cases are quieter.

A caseworker says, “We are doing a study and I would really appreciate your help.”

A counselor hands a research form to a client immediately before a therapy session.

A program officer asks beneficiaries individually whether they have completed the organization's survey.

No one says services will be withdrawn. Yet a client may reasonably wonder whether cooperation affects how staff view their case.

OHRP emphasizes that influence is contextual and that informed consent must be sought under circumstances minimizing coercion and undue influence.

Researchers therefore need to consider whether participants may experience an invitation as pressure even without an explicit threat.

The person delivering the invitation matters

A client may find it easier to decline an invitation from an independent researcher than from the caseworker who decides whether documentation is complete, the counselor with whom they have built a therapeutic relationship, or the staff member who regularly helps them access services.

This does not mean service providers must never mention research.

It means recruitment design should ask whether the professional relationship makes refusal unnecessarily difficult and whether another person or channel could handle recruitment or consent.

Where independent recruitment is not feasible or appropriate, other safeguards may reduce pressure. The exact arrangement should be reviewed under the ethics framework governing the study.

Clients need to know what will happen if they say no

“Participation is voluntary” is abstract.

Potential participants may need a more concrete explanation relevant to their circumstances.

Will counseling continue? Will the housing application still be processed? Will benefits remain unchanged? Will the person's place in a program be affected? Will their caseworker know they declined?

OHRP's guidance on nonfinancial incentives states that choosing not to participate should not adversely affect the person's relationship with the institution or its staff or the provision of services, and it identifies threatened loss of services or program access to which someone is otherwise entitled as overt coercion.

Where these concerns are foreseeable, specific assurances may communicate voluntariness better than generic language alone.

Participation information itself may be sensitive

Suppose an independent researcher conducts confidential interviews, but the service provider receives a list of clients who accepted or declined.

The interview answers may be protected while the research decision remains visible.

For clients dependent on staff, that can matter.

A person may worry that a caseworker will view refusal as uncooperative, ungrateful, suspicious, or uninterested in the program. Researchers should therefore ask whether service providers genuinely need to know individual participation decisions.

Where feasible and appropriate, recruitment may be structured so that staff responsible for services do not receive unnecessary information about who accepts, declines, or withdraws.

Research records and service records should not silently merge

Research conducted inside a service organization can blur information systems as well as relationships.

Will research responses enter the client's ordinary case file? Can service staff see them? Can the information affect eligibility or service planning? Will researchers extract information from existing records? Can identifiable research data later be used for administrative purposes?

These questions should be addressed before recruitment and accurately reflected in the approved consent process.

Researchers should not promise that research is “completely separate” from services if organizational systems or study procedures do not actually maintain that separation.

Confidentiality may be particularly important when participants criticize the service

Many studies involving clients ask precisely the questions organizations most need answered:

Was the service helpful? Were staff respectful? Did you experience discrimination? What went wrong? Did you feel safe? Would you complain about anything?

Those questions can produce useful evidence while also creating obvious concerns about candor.

Clients may reasonably hesitate if they believe critical responses can reach the people responsible for their continuing services.

Researchers should therefore consider data access, reporting thresholds, de-identification, small subgroup risks, quotations, and other ways identities might become apparent.

A survey can omit names and still reveal a respondent when only one person fits the description.

Offering extra services can create undue influence

Threatening to remove an existing service is not the only concern.

A study might offer access to an additional program, priority appointment, special resource, or other nonfinancial benefit available only to participants.

OHRP recognizes that access to services or programs can constitute a nonfinancial incentive and may create undue influence in some circumstances.

This does not mean research can never offer a benefit. Some studies are specifically evaluating interventions or programs that participants receive through the research.

The ethical question is whether the offer is accurately described and whether its value and conditions improperly compromise voluntary choice. Researchers should distinguish an attractive offer from coercion while still evaluating possible undue influence.

Research-only services need to be distinguished from ordinary entitlements

Suppose a study tests a new counseling program available only within the research protocol.

Declining the research may mean the person does not receive that experimental research program. That is not necessarily a threatened loss of an existing service to which the person is otherwise entitled.

The distinction should be communicated carefully.

Potential participants need to know what services remain available outside the research, what is available only through the study, and whether any ordinary services change if they decline.

Researchers should avoid language suggesting that the organization will abandon people who choose standard services rather than research.

Gatekeepers can create pressure on behalf of the research team

Organizations often rely on directors, caseworkers, community leaders, program coordinators, or front-line staff to identify and approach eligible service users.

A researcher may prepare an ethically appropriate invitation, only for a staff member to say, “Please participate. We need everyone to support this project.”

The resulting pressure may originate with the intermediary rather than the investigator.

Recruitment planning should therefore address how gatekeepers communicate the study and what information they receive about participation.

Some service users may face several vulnerabilities at once

People using social or community services may also experience economic hardship, unstable housing, disability, migration-related vulnerability, limited access to healthcare, or other circumstances that make particular services especially important.

Researchers should avoid treating such populations either as incapable of autonomous choice or as ethically interchangeable with participants who have abundant alternatives.

CIOMS emphasizes that vulnerability should be considered contextually and that particular circumstances may threaten a person's capacity to protect their own interests without automatically eliminating their ability to provide voluntary consent.

The response should be proportionate safeguards, not automatic exclusion or paternalistic assumptions.

When a Satisfaction Survey Comes From the Person Handling Your Case

The organization can still learn from the experiences of people who use its services. The ethical design simply prevents access to those services from becoming the bargaining chip that secures participation.

Common Mistakes When Recruiting Clients and Service Users

Separate Research Cooperation From Service Eligibility

If potential participants already receive something from you or your organization, start by identifying everything they may believe you control.

Do not limit the analysis to formal eligibility. Consider waiting lists, scheduling, referrals, case recommendations, staff discretion, documentation, future programs, and the quality of the ongoing professional relationship.

One practical test is particularly useful: imagine that the client says, “No, I don't want to participate.” What changes tomorrow?

If the correct answer is “nothing about the ordinary services or relationship that should depend on this decision,” make sure the research procedures support that answer rather than merely stating it.

Before Recruiting Your Own Clients or Service Users

Questions About Recruiting Clients and Service Users

Clients Should Never Have to Purchase Services With Research Participation

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.