How Do You Know Whether a Problem Requires New Research or Better Use of Existing Knowledge?

Does This Problem Actually Need Another Study?

You identify an important problem, find people affected by it, and begin thinking about a study. The instinct seems reasonable: if the problem persists, perhaps more research is needed.

But the continued existence of a problem does not necessarily mean that knowledge about it is missing. Sometimes researchers already know quite a lot. The difficulty may instead be that the available evidence has not reached the people who need it, has not been implemented consistently, has not been synthesized into a usable form, or does not fit the local context.

This distinction matters because identifying a problem worth investigating is not quite the same as establishing that another primary study is the appropriate response. Before asking, “What study should I conduct?” there is a prior question: “What exactly is missing here?”

New Research Is Needed When Relevant Knowledge Is Genuinely Insufficient

The Critical Question Is What Kind of Gap Actually Exists

A real-world problem and a knowledge problem are not the same thing

Suppose university students continue to struggle with a particular academic skill. That is a real problem. It does not automatically follow that researchers lack knowledge about how to improve that skill.

Perhaps several well-conducted studies have already identified effective approaches. The persistent difficulty could instead arise because those approaches are rarely implemented, instructors lack resources to use them, institutional policies create barriers, or practitioners are simply unaware of the evidence.

In such a case, the practical problem remains real even though the underlying knowledge problem may be much smaller than it initially appears. This is why evidence can sometimes show that a research problem is smaller than you first thought.

The reverse can also occur. A familiar problem may have been discussed for years while the evidence needed to understand or address it remains surprisingly weak. Persistence alone tells you little about which situation you are facing.

Ask what is missing before deciding what to produce

A useful way to diagnose the situation is to separate several possibilities that can look similar at first.

These categories can overlap. An implementation problem, for example, may itself create a legitimate research question if there is substantial uncertainty about why implementation is failing or which strategy would work in a particular context.

Existing literature is not the same as sufficient knowledge

Finding many publications does not establish that a question has been answered. Ten weak studies may leave more uncertainty than one rigorous study addressing the right question.

You therefore need to examine the evidence, not merely count it. Consider whether previous studies directly address your population, phenomenon, intervention, exposure, outcome, or context; whether their methods are adequate; whether findings are reasonably consistent; and whether important uncertainties remain.

This is especially important when the literature contains conflicting evidence that may itself constitute a research problem. Apparently contradictory findings might justify additional research, but they might also be explainable through differences in study design, measurement, populations, contexts, or risk of bias. Sometimes careful synthesis can clarify the disagreement without immediately collecting new data.

Searching for a gap is not enough

Researchers sometimes approach the literature with an implicit objective: find something that has not been studied so that a new study can be justified. That reverses the logic.

The stronger question is not simply, “Has anyone studied this exact combination before?” It is, “What consequential uncertainty remains after considering the best available evidence?”

Minor variations are easy to manufacture. A different university, age group, city, platform, profession, or demographic subgroup can make a study technically different from previous work. But difference is not automatically contribution.

Systematically considering previous research helps prevent unnecessary studies

The idea of evidence-based research argues that decisions about new studies should be informed by a systematic and transparent assessment of relevant prior research. Lund and colleagues have argued that new research should build explicitly on existing evidence rather than relying on selective references to earlier studies. Meta-research has also documented inconsistent use of systematic reviews when researchers justify new health studies.

The underlying principle extends beyond any single discipline: before producing more evidence, establish what the current evidence already allows you to conclude and what it still cannot tell you.

How systematic that assessment needs to be depends on the purpose, discipline, stakes, and stage of the project. A student developing an initial proposal may begin with a rigorous literature search and critical appraisal rather than immediately conducting a publishable systematic review. A large clinical trial, by contrast, may require a substantially stronger demonstration that existing evidence leaves consequential uncertainty.

Sometimes the problem is that knowledge exists but is not being used

This possibility is easy to overlook because universities are structured to reward research production. Yet another study is not automatically the best scholarly response to every evidence-practice gap.

Suppose reliable evidence already indicates that a particular teaching practice improves an outcome, but instructors rarely use it. Repeating another study asking whether the practice works may contribute little. More useful work might investigate why adoption is low, whether the practice can be implemented under local constraints, or how implementation can be improved.

Notice that this does not necessarily mean “no research.” It may mean that the research question should move downstream. Instead of asking whether an intervention works at all, you might investigate implementation, adaptation, mechanisms, feasibility, equity, sustainability, or conditions under which it works.

Local evidence can be necessary, but “it has not been studied here” needs justification

Researchers often encounter evidence generated in another country, institution, health system, educational setting, or population. The temptation is to declare a local research gap immediately.

Sometimes that conclusion is defensible. Context can alter mechanisms, feasibility, costs, acceptability, implementation, or outcomes. But local replication should be justified by a plausible reason why the existing evidence may not transfer adequately.

If nothing consequential is expected to differ, geographical novelty alone may be a weak rationale. If meaningful contextual differences exist, however, research on transferability or adaptation may be valuable.

Some problems require decisions, not discoveries

Research reduces uncertainty. It does not eliminate the need for judgment.

An institution might already know that a particular program produces benefits, for example, while still needing to decide whether those benefits justify the financial cost. That decision may depend on priorities, available resources, competing needs, feasibility, and values. More studies showing the same effect may not resolve the disagreement.

This becomes particularly important when a practical problem cannot ultimately be resolved by research alone. Evidence can inform a decision without determining what decision should be made.

New research should have a reasonable chance of changing what we know or do

Uncertainty by itself is not always enough. Some unanswered questions are trivial, impossible to investigate adequately, or unlikely to influence theory, practice, policy, or subsequent research.

A stronger justification connects the remaining uncertainty to consequences. If the answer could reasonably change understanding, decisions, interventions, policies, methods, or future research priorities, collecting new evidence becomes easier to defend.

This also helps distinguish a genuinely useful investigation from an answerable study aimed at the wrong underlying research problem. Methodological feasibility matters, but feasibility alone does not establish that a study should be conducted.

When Another Effectiveness Study Is Not the Most Useful Next Step

The practical problem did not disappear when the literature was reviewed. What changed was the diagnosis of the knowledge problem. The researcher moved from unnecessarily re-asking a relatively well-answered question to investigating an uncertainty that mattered locally.

This is also why a real-world problem has to be translated into something research can actually investigate. The first researchable question you can formulate is not necessarily the question that most needs answering.

Common Mistakes When Justifying Another Study

Decide What Is Missing Before Designing the Study

Before committing to data collection, write down the decision or understanding your proposed study is intended to improve. Then ask what prevents that decision or understanding from being adequately supported now.

You may discover that you genuinely need new observations. You may instead discover that the evidence exists but is fragmented, inaccessible, poorly communicated, difficult to apply, or ignored. Those are different problems and call for different responses.

Importantly, your conclusion can change as you learn more. A proposed project may begin with a plausible knowledge gap and later encounter evidence that substantially answers it. Allowing the research problem to change as your understanding develops is often a sign of careful scholarship rather than poor planning.

Before Concluding That New Research Is Necessary

Questions About Deciding Whether More Research Is Needed

Research the Uncertainty, Not Merely the Existence of the Problem

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.