Should You Change the Research Question Because the Ideal Population Is Inaccessible?

Your research question matters, but the people needed to answer it are out of reach

You may have developed a strong research question only to discover that the ideal participants cannot realistically be recruited. Perhaps the required institution will not provide access, the population is too dispersed, a gatekeeper controls recruitment, or the people you need cannot be reached within your timeline and resources.

At that point, changing the question can feel methodologically backwards. Researchers are usually taught to let the research question determine the method and population, not to rewrite the question simply because another group is easier to recruit.

That principle remains important. Yet feasibility is also part of research design. A scientifically interesting question that cannot be investigated with the evidence you can realistically obtain is not yet a workable study.

The decision is therefore not simply whether to keep or abandon the question. You need to determine how much can change before you are no longer studying the phenomenon you originally intended to investigate.

Try to preserve the scientific question before changing it

Change the recruitment problem before changing the scientific problem, when you can

First identify what the research question actually depends on

Before deciding whether another population can substitute for the ideal one, identify why the original population was chosen.

Was it selected because its members possess the experience being studied? Because the intervention applies specifically to them? Because the theory concerns a particular developmental, occupational, clinical, cultural, or institutional context? Because the phenomenon occurs only under certain conditions?

Those reasons determine how much flexibility you have.

If the population is merely one convenient setting in which a general phenomenon can be observed, another setting may preserve the core question. If membership in that population is conceptually central to the phenomenon, substitution may produce a different study.

Separate the scientific population from your preferred recruitment site

A common source of unnecessary redesign is confusing the population with the site.

Suppose your question concerns the experiences of first-year engineering students using generative AI for programming assignments. You originally planned to recruit from University A, but University A declines access. If comparable first-year engineering students at Universities B and C satisfy the scientific requirements, losing University A may be a site problem rather than a population problem.

Changing sites can sometimes preserve the question almost entirely. Changing the scientific population deserves much closer scrutiny.

Try alternative routes to the same population first

If one recruitment route fails, ask whether the same target population can be reached legitimately elsewhere. Additional institutions, professional organizations, community partners, relevant online spaces, or other appropriate channels may provide access without changing who the study is about.

When the population exists but your current route cannot reach it, the first task is to determine whether the problem belongs to the population itself or merely to your initial recruitment strategy.

This distinction prevents a common overreaction: changing the scientific question because one convenient site said no.

Sometimes the design can change while the underlying question survives

Participant recruitment is not the only possible source of evidence for every research question. Depending on what you are asking, existing datasets, administrative records, documents, archives, publicly available materials, or other sources may sometimes provide relevant evidence.

This possibility must be evaluated against the question itself. A question about participants' lived experiences generally cannot be answered by replacing interviews with administrative records merely because the records are easier to obtain. Conversely, a question about patterns already captured reliably in an existing dataset may not require new participant recruitment at all.

The principle is simple: change the evidence source only when the alternative still measures or illuminates what the question requires.

Substituting a more accessible population can quietly change the study

Suppose your original question asks how emergency-department nurses make decisions when using an AI triage system. You cannot access emergency departments, but nursing students at your institution are easy to recruit.

Nursing students may be interesting participants for another question. They do not automatically substitute for practicing emergency nurses making real clinical decisions. Their professional responsibility, experience, working conditions, exposure to the system, and consequences of decisions differ.

Calling both groups “nurses or future nurses” does not remove those differences.

The central question becomes when using a more accessible population changes the scientific question too much. That judgment should be based on the logic of the study, not simply on recruitment convenience.

Ask what claims the revised sample would permit

A useful way to evaluate a substitution is to work backward from the conclusion you hope to draw.

If you recruit Population B instead of Population A, would you still be justified in making the same substantive claim? Or would you need to rewrite the conclusion?

If the same conclusion would no longer be defensible, the change is not merely logistical. The scientific question has moved.

Do not preserve the wording while changing what the study actually investigates

A particularly troublesome compromise occurs when researchers change the participants but leave the original research question untouched.

For example, a question about “experienced secondary-school teachers” remains written exactly that way while the sample gradually expands to include preservice teachers because they are easier to recruit. The methods and question now refer to different populations.

If the accessible population changes the phenomenon being investigated, revise the question explicitly. Methodological coherence is more important than preserving an attractive sentence from the original proposal.

Feasibility can legitimately shape a research question

Research questions do not emerge independently of what can be studied. Time, measurement, ethics, data availability, participant access, and resources all constrain what evidence can realistically be obtained.

A feasible question is not necessarily a less ambitious question. It is one whose scope is aligned with the evidence that the researcher can responsibly obtain. This is part of assessing whether a research question is feasible rather than merely interesting.

The danger lies at both extremes. One extreme is refusing to revise an impossible question. The other is allowing convenience to determine the question so completely that the study loses its original scientific rationale.

Changing the question may be better than pretending the substitute population is equivalent

If no realistic route reaches the ideal population and a more accessible population would meaningfully alter the study, explicitly changing the research question may be the more rigorous choice.

You might narrow the question to the population you can genuinely study, investigate a related mechanism, conduct an exploratory study in the accessible group, or frame the work as preliminary evidence rather than making claims about the inaccessible population.

What matters is transparency between the question, population, evidence, and intended inference.

When losing the ideal site does not require abandoning the question

A feasible substitute is not automatically a scientifically equivalent substitute

Change only as much of the study as the access problem requires

When the ideal population becomes inaccessible, move from the least scientifically disruptive response to the most substantial redesign. This helps preserve the original purpose without pretending that every substitution is equivalent.

The objective is not to protect the original wording at all costs. It is to protect alignment among the research question, population, evidence, and conclusions.

Before changing the research question, check what actually needs to change

Questions about revising a study when the ideal population is inaccessible

Preserve the question when you can, revise it when the available evidence requires you to

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.