What Should You Do When the Practical Problem Is Real but Research Cannot Resolve It?

What if the problem matters but producing more evidence will not solve it?

A university knows that many students cannot afford the devices required for online learning. A hospital knows that demand exceeds the number of available beds. A community knows that flooding repeatedly damages homes. An organization knows that salaries are uncompetitive but does not have the budget to increase them.

These problems are real. They may be urgent. They may also be documented exceptionally well.

Yet another study does not automatically make the constraint disappear.

This creates an important distinction for researchers. Some practical problems persist because relevant knowledge is missing. Others persist because action requires money, authority, infrastructure, political agreement, organizational change, value judgments, or difficult trade-offs. Research may inform those decisions, but it cannot make them on behalf of the people or institutions responsible.

The challenge is to determine whether there is still a meaningful uncertainty that research can address without pretending that answering it will resolve the entire practical problem.

Identify what research can change and what it cannot

Not every important practical problem is primarily a knowledge problem

Research produces knowledge, not every condition required for action

Research can estimate prevalence, investigate relationships, evaluate interventions, examine experiences, test explanations, model scenarios, study implementation, and reduce uncertainty in many other ways. These contributions can be highly consequential.

But practical action often depends on conditions that evidence cannot supply.

The distinction matters because researchers can otherwise diagnose every practical difficulty as a shortage of research. Sometimes the problem is not that nobody knows enough. It is that acting on what is already known is difficult.

Ask whether uncertainty is actually preventing action

Suppose a hospital experiences chronic understaffing. Administrators already have reliable workforce records showing vacancy levels, overtime, turnover, patient demand, and staffing requirements. Staff consistently identify compensation and workload as major concerns. Management agrees that additional positions are needed but cannot hire because of a binding budget ceiling.

A new survey documenting that employees feel understaffed may produce publishable data, but it does not necessarily address the constraint preventing action.

Before initiating another study, ask a counterfactual question: if you received a perfectly clear answer tomorrow, would the people responsible be able to act differently?

If the answer is yes, research may address a decision-relevant uncertainty. If the answer is no, identify what else is blocking action.

A problem can be researchable without research being sufficient to solve it

This distinction is easy to miss.

Urban flooding, educational inequality, antimicrobial resistance, poverty, misinformation, employee turnover, and climate-related risks can all generate researchable questions. Yet no single study, and often no body of research by itself, can eliminate the practical problem.

Research might identify mechanisms, evaluate interventions, forecast outcomes, or reveal how different populations are affected. Action may subsequently require governance, funding, regulation, infrastructure, professional practice, behavioral change, or coordination among institutions.

A problem therefore does not become unsuitable for research simply because research cannot solve it completely. The researcher needs to specify the contribution more modestly.

Some questions are partly empirical and partly normative

Research can tell decision-makers what is likely to happen under different options. It cannot necessarily determine which outcome society ought to prefer.

Imagine that a university has evidence that requiring students to purchase a particular technology would improve access to certain learning activities but impose substantial costs on financially disadvantaged students. Research can estimate costs, document experiences, compare alternatives, and evaluate educational outcomes.

It cannot establish, through empirical analysis alone, how the institution should balance educational benefits against affordability and equity. That requires judgments about values and priorities.

The two can inform each other, but they should not be collapsed. Evidence can make a decision better informed without making the decision automatic.

The practical problem may contain a smaller researchable uncertainty

When the entire problem cannot be resolved through research, look for a consequential uncertainty within it.

A city may not be able to eliminate traffic congestion, but research might estimate how different scheduling policies affect peak demand. A university may not be able to eliminate student poverty, but it could investigate which forms of institutional assistance reach students most effectively. A hospital may face unavoidable resource limits while still needing evidence about how available resources can be allocated more efficiently.

This is similar to turning a broad real-world problem into something research can actually investigate. The difference is that you explicitly acknowledge that the resulting study contributes evidence rather than promising resolution of the larger problem.

Sometimes the next step is implementation rather than another explanatory study

A practical problem may persist despite substantial evidence about what should be done. In such cases, repeating studies that establish the same basic problem may have diminishing value.

There may still be researchable questions about implementation: why an evidence-supported intervention is not adopted, how it can be adapted to context, whether implementation is equitable, or what barriers prevent sustained use.

But that is a different research problem. The researcher should not continue presenting an already well-established practical condition as though its existence remains uncertain.

This is where it becomes important to ask whether the situation requires genuinely new research or better use of knowledge that already exists.

Stakeholders may expect research to do more than it can

Applied researchers are sometimes asked to “solve” a problem through a study. That expectation deserves clarification at the beginning.

A research team can investigate why a program has low participation. It cannot guarantee that administrators will change the program. Researchers can evaluate competing policies. They cannot determine how elected officials should reconcile competing political priorities. They can identify barriers to access. They cannot manufacture resources that an institution does not possess.

The distinction between the practical objective and the empirical contribution should therefore be explicit when the problem stakeholders want solved differs from the problem research can address.

Stopping can be an intellectually responsible decision

Researchers are trained to formulate questions, collect data, and produce new knowledge. Consequently, “we do not need another study of this particular question” can feel strangely unacademic. There is, after all, always another limitation section waiting to become a grant proposal.

But if the relevant uncertainty has already been addressed adequately, more research may simply reproduce knowledge without changing the practical decision.

The appropriate next step might instead be implementation, policy deliberation, resource allocation, professional action, communication, or evaluation of how existing evidence is being used.

When the missing ingredient is not another study

The practical problem remains real throughout. What changes is the researcher's understanding of where additional evidence could make a difference.

Common mistakes when research cannot solve the whole problem

Find the point where additional knowledge could actually change something

Start by stating the practical problem without converting it immediately into a research problem. Then identify what is preventing resolution.

Is there genuine uncertainty about what is happening? Are the causes unclear? Is the effectiveness of available interventions unknown? Or is the relevant knowledge already adequate while action is constrained by money, authority, capacity, implementation, politics, or values?

Then ask what would change if your study answered its proposed question convincingly. This is a useful test of decision relevance.

A practical problem does not have to become a research project. When research is appropriate, define its contribution in terms of what it can realistically clarify, evaluate, estimate, explain, or inform.

Before launching another study of an unresolved practical problem

Questions about practical problems that research cannot resolve alone

Do not confuse a need for action with a need for another study

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.