Should You Contact Potential Sites or Data Providers Before Finalizing Your Research Question?

Should You Ask About Access Before Your Research Question Is Final?

You have a promising research question. The participants are at a particular hospital, school, university, company, or community organization. Perhaps the records you need belong to a government agency, research repository, platform, laboratory, or private organization.

There is only one problem: you do not yet know whether that site or provider will actually let you conduct the study or obtain the data.

Researchers sometimes postpone this question because they assume access should be pursued only after the proposal, protocol, or research question is finalized. In some settings, formal permission does indeed require a developed protocol, ethics approval, institutional endorsement, or other documentation. But that does not necessarily mean you should build an entire project around an untested assumption that access will eventually be granted.

The practical issue is whether some form of preliminary feasibility contact is appropriate before the study becomes difficult to change.

Contact Them Early When the Study Critically Depends on Their Cooperation

What Early Contact Can Tell You Before You Commit

Some research questions contain hidden access assumptions

A question can look perfectly feasible on paper because the access problem is buried inside the proposed method.

Consider a question about burnout among intensive-care nurses at three hospitals. The question implicitly assumes that those hospitals will permit recruitment. A study using ten years of student records assumes that the institution possesses the required variables and can release them for the proposed research. A project analyzing proprietary platform data assumes that the provider offers a route through which those data can actually be obtained.

Until those assumptions are investigated, they remain assumptions.

This matters because access can determine whether the research question is feasible. If losing one site or dataset would make the study impossible, access is not a minor administrative detail. It is part of the study's feasibility architecture.

Preliminary feasibility contact is different from formal research permission

One of the most important distinctions is between asking whether a proposed project might be possible and receiving authorization to conduct it.

A department head saying, "This sounds interesting; send us the proposal when it is ready" may be useful evidence that further discussion is possible. It is not necessarily permission to recruit employees. A data manager confirming that certain records exist does not necessarily authorize their release. A collaborator's willingness to help does not necessarily bind the institution the collaborator works for.

Keep these levels of commitment separate in your planning and documentation.

Early contact can tell you whether the required population is actually there

If your study depends on participants at a particular site, preliminary contact can help test assumptions about the available population.

You might discover that the organization has far fewer eligible participants than expected, that the relevant program has recently closed, that participants are distributed across branches you had not considered, or that another study is already recruiting heavily from the same group.

You may also learn that the institution cannot provide exact counts before formal approval but can offer enough information to judge whether the proposed recruitment target is plausible.

This is particularly valuable when determining whether you can recruit the participants your study requires. Population figures from websites, annual reports, or general institutional statistics may bear little resemblance to the number satisfying your complete eligibility criteria.

Early contact can reveal whether the data you need actually exist

Researchers sometimes design secondary-data studies around variables they assume an organization must collect.

The assumption may be wrong.

A university may record enrollment but not the specific learning-behavior variables your question requires. A hospital may have electronic records but not consistent historical coding for the period you want to study. An organization may collect a variable only after a particular year. Records may exist in separate systems that cannot readily be linked. Historical data may have been archived, aggregated, or deleted under retention policies.

A preliminary enquiry can therefore establish whether it is worth investigating whether the data required by your study actually exist before you build the methodology around them.

Existence is only the first question. Even if the data are present, their format, completeness, coverage, definitions, granularity, documentation, and accessibility may determine whether they can answer your question.

The person you first contact may not be the person who can grant access

Researchers sometimes obtain enthusiastic support from a colleague inside an organization and assume that institutional access has been secured.

That colleague may have no authority to approve the study.

Depending on the setting, authorization might involve a research office, school administrator, hospital research committee, department chair, data protection office, records custodian, information-security team, legal office, ethics committee, institutional review board, database owner, community authority, or some combination of these.

Early contact should therefore identify not only whether someone is interested but who actually controls the access you need and what process that person or office requires.

Different kinds of access may require different permissions

"Permission to conduct the study" can conceal several distinct forms of access.

A school might permit researchers to use a classroom without giving them access to student records. A hospital might allow recruitment but prohibit the release of identifiable patient information. A company might distribute a survey invitation to employees while refusing to provide employee email addresses. A data repository might approve access to de-identified records but not permit linkage with another dataset.

For HHS-conducted or HHS-supported non-exempt human-subjects research in the United States, Office for Human Research Protections guidance illustrates how institutional involvement can differ depending on what an institution actually does. For example, merely allowing outside investigators to use facilities or informing potential participants about a study may be treated differently from obtaining informed consent, interacting with participants for research purposes, or obtaining identifiable private information.

The regulatory details vary by jurisdiction and study. The broader planning lesson is portable: specify exactly what cooperation you need instead of asking vaguely whether an organization will "allow research."

Ask operational questions, not just whether the organization is interested

An encouraging "yes" can be surprisingly uninformative.

A better preliminary discussion investigates the conditions under which the project could actually proceed. Depending on the study, useful questions may concern the approximate eligible population, available data fields, historical coverage, recruitment procedures, approval sequence, expected review time, fees, required agreements, permitted data uses, security requirements, available facilities, staff involvement, and restrictions on publication or data sharing.

For a participant-based study, you might need to know whether investigators can approach participants directly or whether the organization must distribute invitations. For a data-based project, you might need to know whether individual-level records can leave the organization's systems, whether only aggregated outputs can be released, or whether analysis must occur in a controlled environment.

These details can change the study design even when access is theoretically possible.

Ask what documentation will eventually be required

Formal access may depend on documents that take time to prepare or obtain.

These could include a complete protocol, institutional endorsement, ethics approval, data-management plan, confidentiality agreement, data-use agreement, information-security review, researcher credentials, training certificates, proof of funding, insurance, collaboration agreement, or other materials specific to the organization.

Do not assume that every organization uses the same sequence. One provider may review a data request only after ethics approval, while another may require a provisional access decision before an ethics application can be completed. Multisite studies may add further institutional arrangements.

The useful question is therefore not simply "Will you give me access?" but "What process would have to occur before access could be granted?"

Ask how long the access process normally takes

Access that arrives after your thesis deadline is functionally no access at all.

An organization may be willing to cooperate but require several rounds of review, contracting, ethics documentation, information-security assessment, or administrative approval. Data extraction itself may enter a technical queue. A research site may accept applications only at particular times of year.

Ask for a realistic indication of the process and typical timeline where the organization can provide one. Treat it as an estimate rather than a guarantee, especially when several offices are involved.

This information should feed directly into whether the study can be completed within your deadline.

Do not start recruiting or collecting data simply because you made contact

Preliminary feasibility enquiries should remain preliminary.

Contacting an administrator to ask whether a population is present is different from approaching individuals as research participants. Asking a data custodian whether certain variables exist is different from receiving or analyzing those data. Discussing possible collaboration is different from beginning research procedures.

Applicable ethics, privacy, institutional, contractual, and regulatory requirements vary. Researchers should follow the requirements of their own institution and the organizations involved before beginning activities that constitute recruitment, consent, data collection, access to identifiable information, or other regulated research procedures.

OHRP guidance provides a useful illustration of why these distinctions matter. Under the HHS framework to which that guidance applies, obtaining informed consent, interacting with participants for research purposes, or obtaining identifiable private information can make an institution engaged in human-subjects research, whereas some more limited activities, such as informing prospective participants about a study or permitting outside investigators to use facilities, may not.

Informal enthusiasm is useful information, not guaranteed access

Imagine that a hospital administrator responds, "We would be happy to support this project." That is encouraging. Six months later, the formal review may still identify privacy, staffing, legal, data-security, or recruitment concerns that prevent the study from proceeding as originally proposed.

The safest planning language reflects the actual status. You might have a potential site, an interested collaborator, preliminary institutional support, an application under review, conditional approval, or formal authorization. These are not interchangeable.

This distinction becomes especially important when your study depends on data controlled by someone else's permission.

Critical dependencies deserve earlier verification

Not every possible research site needs to be contacted while you are still brainstorming. Early contact is most valuable when the proposed project contains a dependency whose failure would force substantial redesign.

If you have ten plausible recruitment sites and need only two, uncertainty about one site may be manageable. If your entire dissertation requires a proprietary dataset held by one organization, that organization's access policy is much more consequential.

A useful way to prioritize preliminary enquiries is to ask: If this organization says no, what happens to the study?

If the answer is "almost nothing," early confirmation may not be urgent. If the answer is "the research question can no longer be answered as designed," you have identified something worth investigating before the project becomes fixed.

Sometimes you cannot obtain a definitive answer early

Organizations may reasonably refuse to commit before seeing a full proposal. Data custodians may need a detailed variable list. Sites may require ethics approval before considering recruitment. Some access decisions may depend on circumstances that cannot be known months in advance.

In those cases, the purpose of preliminary contact is not necessarily to obtain a guaranteed yes. It may be enough to determine that the proposed access route exists, understand the requirements, identify major obstacles, estimate the timeline, and learn what uncertainty remains.

You can then decide whether the unresolved risk is acceptable.

Build a fallback into studies that depend heavily on external cooperation

If feasible, avoid designing a study in which one unconfirmed organization represents a single point of failure.

You might identify alternative recruitment sites, secondary data sources, a narrower version of the research question, or another design that could still produce meaningful evidence if the preferred access route fails.

A fallback should not be an entirely different study improvised after months of delay. Ideally, it is considered while the question is still flexible enough to accommodate reality.

One Email That Changes the Research Question Before It Is Too Late

The early enquiry did not guarantee the study. It did something more immediately useful: it prevented the researcher from spending months refining a design around data that could not exist in the form originally assumed.

Common Mistakes When Seeking Preliminary Access

Verify the Assumptions That Could Break the Study

Look at your proposed research question and identify everything you need from someone else. This might include participants, administrative records, a dataset, laboratory facilities, equipment, organizational cooperation, recruitment distribution, specialist services, or permission to work at a particular site.

Then rank those dependencies by consequence. The more completely the study depends on one external decision, the earlier you should understand the access pathway.

Early contact is therefore not about obtaining premature approval. It is about replacing critical assumptions with information. That distinction becomes especially important before deciding whether to base a project on data you have not yet been guaranteed access to.

What Should You Verify With a Potential Site or Data Provider?

Frequently Asked Questions About Contacting Research Sites and Data Providers

Do Not Build a Study Around an Access Assumption You Could Have Checked

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.