Defining the Population (Study Population)

In a research manuscript, defining the population means stating, before any sampling or data collection is described, exactly who or what the study’s conclusions are meant to apply to. It is a design and reporting step, not a statistical calculation: it answers “what group is this study about,” while sampling (see Justifying Sample Size in a Manuscript) answers the separate question of how many units from that group are actually measured, and how they are selected. A manuscript that moves straight from a research question to a sample size, without an explicit population definition in between, leaves reviewers unable to judge whether the sample is representative of anything in particular.

Target Population vs. Accessible (Study) Population

Research-methods literature distinguishes two layers of “population,” and conflating them is one of the more common methods-section weaknesses:

• Target population — the full group the researcher ultimately wants to describe or generalize to (e.g., “all adults with type 2 diabetes”). This is usually broader than what any single study can actually reach.
• Accessible population (also called the study population or source population) — the subset of the target population the researcher can realistically access given the study’s setting, timeframe, and resources (e.g., “adults with type 2 diabetes attending a specific set of clinics over a defined period”). The sample is drawn from the accessible population, not directly from the target population.

The gap between these two layers is exactly where generalizability claims live or die: a study can only support conclusions about its accessible population with full confidence, and extending those conclusions to the broader target population is an inferential step the discussion section should justify explicitly, not assume.

Inclusion and Exclusion Criteria

A population definition is operational only once it specifies who qualifies and who does not:

• Inclusion criteria — the characteristics a unit (person, document, organization, sample, event) must have to be considered part of the population at all.
• Exclusion criteria — characteristics that disqualify an otherwise-eligible unit, applied after inclusion criteria are met (for example, meeting a diagnosis-based inclusion criterion but being excluded for a comorbidity that would confound the outcome).

Reporting guidelines used across clinical and observational research treat this as a standalone, mandatory reporting item, separate from the sample-size justification that follows it. CONSORT 2010 (for randomized trials) requires eligibility criteria for participants as item 4a; STROBE (for cohort, case-control, and cross-sectional designs) requires, as item 6, the eligibility criteria and the sources and methods of participant selection — the observational-design version explicitly asks for more than the criteria alone, because how units were found and selected also affects what the resulting sample can represent.

Why This Is Not the Same as Sampling

Defining the population and drawing a sample are sequential, distinct steps, and a manuscript should keep them visibly separate rather than folding the population definition into a single sentence about sample size:

• The population is the group being described or generalized to — it can be enumerable (e.g., “all registered clinical trials on a given topic”) or, more often in human-subjects research, too large to study in full.
• The sample is the specific subset of the accessible population that is actually measured, selected through a stated sampling method (random, convenience, purposive, stratified, and so on).
• Population characteristics are described with parameters; sample characteristics are described with statistics used to estimate those parameters — a distinction that matters for which inferential methods are appropriate later in the methods and results sections.

A methods section that only reports a sample size and a sampling method, without first stating the population that sample was drawn from and how eligibility was determined, gives a reader a number without the context needed to judge what that number generalizes to.

Applying This Outside Clinical Research

The same logic applies beyond clinical trials. A bibliometric or scientometric study defining its population as records in a database, or a policy study defining its population as institutions meeting a funding threshold, still needs the same three elements — a stated target vs. accessible boundary (or an explicit note that they coincide), inclusion criteria, and exclusion criteria — for the same reason: so another researcher can assess or attempt to reproduce exactly which units were eligible for the study.

Frequently Asked Questions

Is “study population” the same as “sample”?

No. The study population (or accessible population) is the bounded group a sample is drawn from; the sample is the specific subset actually measured. A manuscript should define the population before describing how the sample was selected from it.

Where does the population definition belong in a manuscript?

In the methods section, typically at or near the start of a “Participants” or “Study Population” subsection, before the sampling method and sample-size justification, and before any statistical analysis plan.

Do inclusion and exclusion criteria need separate justification?

Reporting guidelines such as CONSORT and STROBE expect eligibility criteria to be stated as their own checklist item; when a criterion is not self-evidently relevant to the research question (for example, excluding a comorbidity), a brief rationale strengthens the methods section by showing the boundary was a deliberate design decision, not an arbitrary or post hoc one.