Which Research Methods Are Missing From the Existing Literature?

What Can the Literature Not Investigate Well With the Methods It Currently Uses?

Once you identify the methods that dominate a literature, the obvious next question is what is missing. Perhaps almost every study is cross-sectional and none follows participants over time. Perhaps researchers rely on self-report while behavioral evidence is scarce. Perhaps experimental studies establish effects but little qualitative work examines how participants experience the intervention.

These patterns can reveal methodological gaps. They can also tempt researchers into weak reasoning: “Nobody has used Method X, so I will.”

A method does not become scientifically necessary merely because it is absent. A meaningful methodological gap exists when the available approaches leave an important research question, inference, mechanism, population, outcome, or evidential limitation insufficiently addressed and another methodological approach could help resolve it.

Identify Missing Evidential Capabilities Before Missing Method Names

How to Find Methodological Gaps That Actually Matter

Start With the Unresolved Question, Not Your Preferred Method

Methodological gap hunting often goes backward. A researcher wants to conduct interviews, structural equation modeling, an experiment, a longitudinal study, or a mixed methods project and then searches for evidence that nobody has used that approach before.

A stronger sequence begins with the knowledge problem.

This keeps the research question in charge of the methodology rather than the other way around.

Map What the Existing Methods Can and Cannot Support

Begin with the methods that currently dominate the literature. Then ask what forms of inference those approaches are well suited to support.

Different methodological structures provide different kinds of leverage. A cross-sectional design may characterize prevalence or association at a given period but generally offers less information about within-person change or temporal ordering. Longitudinal designs can examine change over time but do not automatically eliminate confounding. Randomized studies can provide strong leverage for estimating intervention effects under appropriate conditions, but some questions cannot feasibly or ethically be randomized.

Cochrane explicitly recognizes that some review questions cannot be answered by randomized trials alone and that non-randomized evidence may be useful for long-term or rare outcomes, different populations and settings, or forms of intervention delivery not adequately represented in randomized trials. It simultaneously cautions that non-randomized intervention studies require careful assessment of confounding and other biases.

There is no universally superior design detached from a question. The gap lies in a mismatch between the evidence required and the evidence currently available.

Missing Methods Can Occur at Several Levels

A methodological gap does not always mean an entire study design is absent. The missing capability may be much more specific.

These are examples, not prescriptions. Each methodological response introduces its own assumptions, biases, costs, and feasibility requirements.

An Absent Method Is Not Automatically a Gap

Suppose no randomized trial exists in a literature. That absence may matter if the central question concerns the causal effect of an intervention that can ethically and feasibly be randomized. But if the question concerns a naturally occurring exposure that researchers cannot assign, the absence of randomization may be entirely unsurprising.

Likewise, a field without ethnography does not automatically need ethnography. A field without machine learning does not automatically need machine learning. A field without mixed methods does not automatically need mixed methods. Academic methodology is not a bingo card.

The distinction is crucial because novelty of technique and contribution to knowledge are not the same thing.

Sometimes the Missing Element Is Time

Many apparent methodological gaps concern temporal structure rather than an entirely different research tradition.

If most studies collect data once, the literature may have substantial evidence about differences and associations but little evidence about development, persistence, sequence, or change. A longitudinal design could then be valuable if the unresolved question genuinely concerns time.

However, simply measuring the same variables twice does not automatically solve the problem. Follow-up timing, attrition, measurement consistency, analytical strategy, confounding, and the substantive process being studied all matter.

Sometimes the Missing Element Is Measurement

A literature can use diverse study designs while repeatedly measuring a construct in the same way. Heavy dependence on self-reported behavior, for example, may leave uncertainty about actual behavior. Reliance on one instrument may make the evidence vulnerable to the assumptions and limitations of that measure.

In such cases, the methodological gap may concern measurement rather than design. Adding behavioral traces, direct observation, performance measures, administrative records, validated alternative instruments, or another relevant source may provide information the existing evidence cannot.

This is especially important when examining outcomes that researchers repeatedly measure. Apparent agreement across studies may partly reflect repeated use of the same operationalization.

Sometimes the Missing Element Is Process or Explanation

A mature quantitative literature may establish that an intervention produces an effect without explaining how participants experience it, how implementation varies, or why it succeeds in some circumstances and not others.

Qualitative research can be valuable for questions concerning experiences, meanings, processes, implementation, and context. Conversely, rich qualitative evidence does not automatically provide estimates of prevalence, average effects, or population distributions.

The methodological opportunity arises when the unresolved question requires the form of evidence that is missing.

Mixed Methods Should Solve an Integration Problem

Researchers sometimes describe mixed methods as inherently more comprehensive. That is too simple.

NIH's Office of Behavioral and Social Sciences Research characterizes mixed methods research as the intentional collection, analysis, and integration of quantitative and qualitative data when combining them can provide a more comprehensive understanding of a research problem than either form alone. The key word is integration. Running a survey and a few interviews side by side does not automatically create a useful mixed methods design.

Use mixed methods when the research question requires complementary evidence and the integration itself contributes to the answer.

A Missing Method May Be Missing for a Good Reason

Before proposing an unfamiliar design, investigate why previous researchers have not used it.

The method may be ethically inappropriate, prohibitively expensive, infeasible for the population, incompatible with the phenomenon, unable to achieve sufficient follow-up, or dependent on data that do not exist. Sometimes the absence reflects disciplinary habit, but sometimes it reflects a genuine constraint.

Different Does Not Necessarily Mean Better

Replacing the dominant method with another method may simply exchange one set of limitations for another.

Cochrane's guidance illustrates this clearly for intervention evidence. Non-randomized designs can answer questions not adequately covered by randomized trials, but they may introduce greater concerns about confounding and selection. Conversely, randomized designs have important strengths for causal estimation but may not cover every relevant population, outcome, setting, or time horizon.

A strong methodological justification therefore states both what the proposed approach adds and what limitations remain.

Connect the Methodological Gap to the Weakness in the Evidence

The strongest justification has a chain of reasoning:

Important question → current evidential limitation → missing methodological capability → appropriate design → information the new study could add.

For example, if existing findings are based almost entirely on cross-sectional associations, and the unresolved question concerns whether changes in one variable precede changes in another, a longitudinal approach has an identifiable purpose. If researchers already have strong longitudinal evidence, adding another longitudinal study merely because one particular statistical model has not been used is a much weaker methodological argument.

This reasoning also helps identify where the evidence remains weak for methodological reasons rather than simply because there are too few publications.

From “Nobody Used Interviews” to a Defensible Methodological Gap

Common Mistakes When Claiming a Methodological Gap

Choose the Missing Method Only After You Know What Evidence Is Missing

Your methodological justification should begin with an unresolved substantive question and end with a design capable of producing the evidence needed to investigate it.

The method should be the consequence of your gap analysis, not the gap itself.

Before Claiming That a Research Method Is Missing

Questions About Missing Methods and Methodological Gaps

The Real Gap Is Missing Evidence, Not an Unused Technique

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.