Should You Narrow a Very Large Literature by Population?

Can You Make an Enormous Literature Manageable by Focusing on One Population?

A search on an otherwise focused topic may still retrieve studies involving children, adolescents, university students, working adults, older adults, patients, professionals, or entire communities. Restricting the review to one group can remove thousands of records almost immediately.

Sometimes that is exactly what the research question requires. A review about first-year university students does not become methodologically suspect because it excludes secondary-school students. In other cases, however, the population restriction appears only after the researcher discovers how many papers must be screened.

The important question is not whether a population restriction makes the literature smaller. It almost certainly will. The question is whether the population characteristic identifies the people to whom your research question actually applies.

Yes, When the Population Boundary Comes From the Question

Define Who the Evidence Is Supposed to Represent

Population Is Part of the Question, Not Merely a Search Filter

In many review frameworks, population is one of the central elements used to define scope. In PICO, for example, the P represents the population or participants. In qualitative evidence synthesis, frameworks such as PICo similarly require explicit attention to the population alongside the phenomenon of interest and context.

Cochrane guidance recommends defining participant eligibility in advance and balancing two competing needs: criteria should be broad enough to capture relevant diversity but narrow enough that combining the studies can produce a meaningful answer. JBI similarly links population eligibility directly to the review question and notes that relevant characteristics can include age, gender, ethnicity, clinical or socioeconomic characteristics, health conditions, and other variables justified by the review objective.

That balance matters. A population can be defined so broadly that the resulting evidence becomes difficult to interpret, but it can also be defined so narrowly that potentially applicable evidence disappears for little substantive reason.

Ask Which Population Differences Could Change the Answer

Not every difference between participants deserves an eligibility boundary. Researchers could divide almost any population by age, geography, occupation, institution type, socioeconomic status, diagnosis, experience level, or dozens of other characteristics. The existence of a category does not establish its relevance.

A useful starting question is: Why might the answer to my research question differ for this group?

For an educational intervention, developmental stage may matter. For a workplace technology, occupational role may affect how the system is used. For a clinical intervention, diagnosis or disease severity may alter effects. For an educational-policy question, institutional level or setting may define the environment in which the policy operates.

Cochrane guidance specifically advises that restrictions based on population characteristics should have a sound rationale. It also distinguishes participant-level characteristics, such as age or disease severity, from study-level characteristics such as care setting or geographical location because these distinctions can affect how evidence is grouped and synthesized.

Define the Population Precisely Enough to Apply the Criterion

“Students,” “adults,” “teachers,” or “patients” may sound like populations, but each can conceal substantial heterogeneity.

If your review concerns university students, does that include undergraduate and postgraduate students? Students in professional schools? Distance learners? People enrolled in continuing education? If it concerns teachers, are university faculty included? What about teaching assistants or clinical educators?

A useful population criterion should be operational enough that two reviewers can apply it consistently. Depending on the topic, you may need to specify age range, educational level, diagnosis, disease severity, occupational role, setting, or another defining feature.

Precision does not mean adding every demographic characteristic imaginable. Include characteristics because they determine relevance to the question, not because they happen to be available in the papers.

Population and Context Can Be Easy to Confuse

Some boundaries describe people. Others describe where those people live, learn, work, or receive services.

“Undergraduate students” is primarily a population. “Universities in rural areas” introduces a contextual or setting boundary. “Nurses working in intensive care units” combines a professional population with a particular setting.

JBI's qualitative-review guidance treats population and context as separate elements precisely because context can shape the relevance and applicability of evidence.

This distinction can help when a literature seems heterogeneous. You may discover that you do not actually need a narrower population. The question may instead concern a specific setting or context.

Do Not Assume Geographic Restriction Is Harmless

A common way to reduce a literature is to include only studies from one country, region, or income classification. Such restrictions can be defensible when geography is integral to the question, perhaps because educational systems, policies, healthcare arrangements, culture, infrastructure, or implementation conditions differ in ways central to the phenomenon.

But geography should not become a convenient proxy for relevance without explanation. Cochrane guidance emphasizes that population and setting restrictions require justification, partly because unnecessary restrictions can reduce the wider relevance of a review.

If the real question is about a particular policy environment, define that environment. If the research genuinely concerns the Philippines, for example, say why evidence from that setting is the object of inquiry rather than merely using national boundaries to reduce retrieval.

Plan for Studies With Mixed Populations

Real studies rarely respect the neat boundaries of a review protocol.

Suppose your review concerns adolescents aged 13 to 17, but a study includes participants aged 12 to 19. Or your population is undergraduate students, while a study combines undergraduate and postgraduate participants. Excluding every mixed sample may discard useful evidence, while including the entire sample may introduce participants outside the intended population.

Cochrane recommends deciding in advance how studies containing only a subset of eligible participants will be handled. Separate data for the eligible subgroup may sometimes be available. When they are not, reviewers need a predefined approach rather than making a different decision for each study after seeing its findings.

Consider Whether Population Differences Require Exclusion or Analysis

A relevant population characteristic does not always need to become an exclusion criterion.

Suppose you expect an intervention to work differently for novice and experienced teachers. One option is to review only novice teachers. Another is to retain both groups and examine experience as a potential source of variation if the review question and available evidence support that analysis.

Cochrane explicitly recognizes this choice: reviewers may narrow the scope by excluding particular subpopulations or maintain broader eligibility and examine important population differences during analysis.

The second option may preserve broader applicability, but it also requires sufficient information and appropriate methods. The correct choice depends on the research objective, not on a general preference for broad or narrow reviews.

Do Not Use Population Restriction to Rescue an Undefined Topic

Imagine a review titled “Effects of artificial intelligence on students.” Restricting it to university students removes schoolchildren, but “artificial intelligence” and “effects” remain extremely broad. The resulting literature may still combine generative AI, predictive analytics, intelligent tutoring systems, assessment technologies, academic achievement, engagement, attitudes, and many other phenomena.

Population is only one dimension of scope. When the literature remains unwieldy, return to the research question before introducing arbitrary restrictions. Depending on what you need to know, it may also be appropriate to specify which study designs can answer the question or which outcomes the review needs to address.

When a Population Boundary Clarifies the Question

The population restriction makes the evidence base smaller, but that is not its methodological justification. The restriction identifies the group about whom the researcher intends to make claims.

Common Mistakes When Population Becomes a Shortcut

Define the People Before You Exclude the Papers

When population appears to be the obvious way to shrink an enormous literature, identify the claims you eventually want to make. Those claims should tell you whom the evidence needs to represent.

The final population should be broad enough to answer a useful question and narrow enough that the answer still means something when the included evidence is considered together. That balance is less glamorous than clicking a demographic filter, but considerably easier to defend.

Before Restricting a Large Literature by Population

Questions About Narrowing Literature by Population

Narrow by Population Only When You Know Whom the Answer Is About

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.