Population vs Sample in Research: Definitions and Examples

Every quantitative study and most qualitative studies make a critical distinction between two groups: the larger group the research is about, and the smaller group the researcher actually collects data from. The first is the population. The second is the sample. Understanding the difference, and knowing how to define both clearly in your methodology section, is one of the foundational skills graduate researchers need to develop. Get it wrong, and your conclusions don't match the evidence you collected. Get it right, and you've established the foundation for everything that follows.

This guide defines both terms with examples, explains why the distinction matters for your conclusions, walks through the major sampling methods, and covers the common mistakes that send manuscripts back for revision. For the broader methodology framework that population and sample sit inside, see our research methodology guide. For the systematic biases that arise when samples don't represent populations well, see our research bias guide.

Quick Answer: Population vs Sample

Population. The complete group of people, organizations, documents, or events your research is about. The population is the group you want to draw conclusions about.

Sample. The smaller subset of the population that you actually collect data from. The sample is the group you observe.

The simplest test. If you're using statistics to generalize beyond the people you measured, those people are your sample, and the broader group you're generalizing to is your population.

Why it matters. Conclusions drawn from a sample only generalize to the population the sample was drawn from. Mismatch between the two is one of the most common reasons reviewers send manuscripts back for revision.

Population vs Sample: At a Glance

The table below summarizes the core differences between population and sample. The detailed explanations and examples follow.

Feature	Population	Sample
Definition	The complete group your research is about	The subset of the population you actually study
Size	Usually large; sometimes very large or even infinite in principle	Always smaller than the population; size depends on study type and goals
Numerical descriptor	Parameter (e.g., population mean, population proportion)	Statistic (e.g., sample mean, sample proportion)
Symbol convention	Greek letters (μ for mean, σ for standard deviation)	Roman letters (x̄ for mean, s for standard deviation)
Knowability	Usually unknown; the goal of the study is to estimate it	Directly observed and measured
Generalization direction	What you want to generalize to	What you observe and use to estimate
Selection logic	Defined first, by the research question	Selected from the population using a sampling method

What Is a Population in Research?

A population is the complete group of people, organizations, documents, events, or other units your research is about. The population is defined by the research question. If your question asks about American adults, your population is American adults. If it asks about peer-reviewed articles published in clinical psychology journals between 2015 and 2025, your population is those articles. The defining feature is that the population includes every member of the group you want to draw conclusions about, whether or not you can actually access them all.

Most populations of research interest are large. American adults numbers more than 250 million people. Peer-reviewed clinical psychology articles published in a decade run into the tens of thousands. International graduate students at US universities number in the hundreds of thousands. The size is part of why we sample. Studying the entire population is rarely feasible, and often impossible.

A useful distinction in research methodology is between the target population (the group the research aims to describe) and the accessible population (the subset of the target population the researcher can actually reach). For example, the target population might be all undergraduate students in the United States, while the accessible population is undergraduate students at the researcher's home university plus three nearby institutions. The methodology section needs to define both, and the limitations section needs to address the gap between them.

Examples of Populations

All American adults aged 18 and over. A typical population for a national survey on financial behavior or political attitudes.
All US households participating in the 2022 Survey of Consumer Finances. A specific institutional population already defined by the survey.
All first-year international graduate students at US R1 universities in academic year 2026 to 2027. A population narrowed by status, institutional type, and time.
All peer-reviewed articles on educational policy implementation published in five top education journals between 2015 and 2025. A document population for a systematic review.
All registered nurses currently practicing in acute-care hospitals in the state of Ohio. A specific occupational population narrowed by setting and geography.

What Is a Sample in Research?

A sample is the subset of the population that you actually collect data from. The sample is what you observe. The conclusions you draw from your data are direct claims about your sample, and indirect (probabilistic) claims about the population the sample was drawn from. Statistical inference is the formal machinery for moving from sample observations to population estimates with quantified uncertainty.

Samples can be large or small depending on the study type. Quantitative survey research often uses samples in the hundreds or thousands. Experiments often use samples in the tens to low hundreds. Qualitative interview studies typically use samples between 12 and 30 participants. Case study research may use a sample of one or a few cases. The right sample size depends on the research question, the analytic approach, and the variability in the population.

A clear methodology section defines the sample with the same rigor as the population. The minimum information includes who is in the sample (defining characteristics), how many (sample size), how they were selected (sampling method), and where they were recruited from (sampling frame). For step-by-step guidance on calculating the right sample size for your study, see our companion article on how to calculate sample size for your study.

Examples of Samples

2,500 American adults recruited through a probability-based online panel. A sample drawn from the population of American adults using a probability sampling method.
485 households included in the public-use file of the 2022 Survey of Consumer Finances. A sample defined by the dataset's release file, drawn from the survey population.
22 first-year international graduate students at one US R1 university recruited through campus listservs and snowball sampling. A purposive qualitative sample drawn from the broader international graduate student population.
147 peer-reviewed articles meeting inclusion criteria after database searching and screening. A sample of articles drawn from a larger population using systematic review procedures.
18 registered nurses recruited from three hospitals in central Ohio. A purposive sample drawn from the broader population of Ohio acute-care nurses.

Population and Sample: Side-by-Side Examples

The same research topic can produce very different population and sample definitions depending on the study design. The examples below show how the two relate in concrete research scenarios.

Research topic	Population	Sample
Financial risk tolerance among US households	All US households	4,602 households in the 2022 Survey of Consumer Finances public-use file
Adjustment experiences of international graduate students	All first-year international graduate students at US R1 universities	22 first-year international graduate students at one US R1 university
Workplace satisfaction in tech companies	All employees of US-based tech companies with 100+ employees	1,500 employees from 12 companies recruited through HR partnerships
Educational policy implementation	All public elementary schools in the state of Ohio	Three schools selected for in-depth case study analysis
Mental health literacy	All adults in the United States	2,000 adults recruited through a probability-based online panel

Notice that in each case, the population is much larger than the sample. The conclusions the researcher can defend are limited by the relationship between the two. A study of three schools in Ohio cannot make claims about all US public schools. A study of 22 graduate students at one university cannot make claims about all international graduate students. The methodology section needs to be honest about what the sample can and cannot support.

How Samples Are Selected: An Overview of Sampling Methods

The sampling method determines whether the sample fairly represents the population. Two broad categories cover most sampling approaches in research.

Probability Sampling

In probability sampling, every member of the population has a known, non-zero chance of being included in the sample. The selection process uses randomization to remove human bias from who gets in. Probability sampling supports formal statistical inference and is the gold standard for quantitative survey research.

Simple random sampling. Every member of the population has an equal chance of selection. The sampling frame (a list of all population members) is essential.
Stratified sampling. The population is divided into subgroups (strata) based on a relevant characteristic, and a random sample is drawn from each stratum. Useful when subgroup comparisons matter.
Cluster sampling. The population is divided into clusters (often geographic), some clusters are randomly selected, and either all members or a sample of members within selected clusters are studied. Useful when populations are spread across many locations.
Systematic sampling. Every kth member of the population is selected after a random starting point. Easier to implement than simple random sampling when a list is available.

Non-Probability Sampling

In non-probability sampling, members of the population don't have known chances of selection. The researcher uses other criteria (convenience, purpose, referral, or quota) to recruit participants. Non-probability sampling is the norm in qualitative research and is also common in quantitative research when probability sampling isn't feasible.

Convenience sampling. Participants are selected based on accessibility (campus students, social media followers, walk-in clients). The most common sampling approach in graduate research and the most vulnerable to selection bias.
Purposive sampling. Participants are deliberately chosen because they have characteristics relevant to the research question. The standard approach in qualitative research, where information richness matters more than statistical representativeness.
Snowball sampling. Existing participants refer additional participants. Useful for hard-to-reach populations but vulnerable to network bias.
Quota sampling. Participants are recruited until predefined subgroup quotas are filled (for example, 50% women, 50% men). Looks like stratified sampling but lacks the random selection within strata.

The choice between probability and non-probability sampling has consequences for the conclusions you can defend. Probability sampling supports formal statistical inference to the population. Non-probability sampling supports description of the sample and analytic generalization (transferability of findings to similar contexts) but not formal statistical inference. Be honest about which kind of sampling your study used and what that means for your conclusions.

Common Mistakes with Population and Sample

The same problems show up in graduate research over and over. Knowing them in advance can save a round of revisions.

Defining the population vaguely or not at all. Methodology sections that say "the population is graduate students" without specifying which graduate students, where, when, or in what programs make it impossible for reviewers to evaluate whether the sample actually represents the population. Be specific.
Generalizing beyond the sampling frame. A study of nursing students at one university can describe those students. It cannot describe all nursing students in the United States. Conclusions need to match the sampling frame, not the researcher's preferred population.
Conflating population and sample. Statements like "the population of this study was 200 graduate students" misuse the term. 200 graduate students is the sample. The population is the larger group those 200 were drawn from.
Ignoring the gap between target and accessible populations. When the accessible population differs from the target population in systematic ways (only one university, only one geographic region, only one time period), the limitations section needs to address what that means for generalizability.
Treating non-probability samples as probability samples. A convenience sample doesn't support formal statistical inference no matter how large it is. Statistical tests that assume probability sampling produce misleading conclusions when applied to convenience data. Acknowledge the sampling method's limits explicitly.
Underpowering the study. Sample size needs to match the analytic approach. Too few participants for the planned analysis means a study that can't detect the effects it was designed to find. For quantitative work, calculate required sample size with a power analysis before data collection begins. For step-by-step guidance, see our sample size calculation guide.

Frequently Asked Questions

What is the difference between a population and a sample in research?

A population is the complete group of people, organizations, documents, or events your research is about. A sample is the subset of the population that you actually collect data from. The population is the group you want to draw conclusions about. The sample is the group you observe. Most research populations are too large to study completely, which is why researchers sample from them. Statistical inference is the formal procedure for moving from observations of a sample to estimates about the population it was drawn from.

Why is the distinction between population and sample important?

Conclusions drawn from a sample only generalize to the population the sample was drawn from. A study of nursing students at one university can describe those students. It can't describe all nursing students in the United States unless the sample was drawn in a way that supports broader inference. Mismatch between the population the researcher claims to be studying and the population the sample actually represents is one of the most common reasons reviewers send manuscripts back for revision.

What is the difference between target population and accessible population?

The target population is the group the research aims to describe (for example, all undergraduate students in the United States). The accessible population is the subset of the target population the researcher can actually reach (for example, undergraduate students at the researcher's home university plus three nearby institutions). The methodology section should define both, and the limitations section should address the gap between them. The gap directly affects how broadly the study's conclusions can be defended.

What are examples of populations in research?

A population can be people, organizations, documents, events, or other units depending on the research question. Examples include all American adults aged 18 and over, all US households participating in the 2022 Survey of Consumer Finances, all first-year international graduate students at US R1 universities in a specific academic year, all peer-reviewed articles published in five top education journals between 2015 and 2025, and all registered nurses currently practicing in acute-care hospitals in a specific state. The defining feature is that the population includes every member of the group the research is about.

What are examples of samples in research?

Sample examples include 2,500 American adults recruited through a probability-based online panel, 4,602 households in a Survey of Consumer Finances public-use file, 22 first-year international graduate students at one US R1 university recruited through campus listservs, 147 peer-reviewed articles meeting inclusion criteria after database searching, and 18 registered nurses recruited from three hospitals in central Ohio. The size and selection method depend on the research question and analytic approach.

What is the difference between a parameter and a statistic?

A parameter is a numerical descriptor of a population (for example, the population mean or population proportion). A statistic is a numerical descriptor of a sample (for example, the sample mean or sample proportion). Parameters are usually unknown; estimating them is the goal of inferential statistics. Statistics are calculated directly from observed sample data and used to estimate the corresponding parameters. Greek letters typically denote parameters (mu for mean, sigma for standard deviation), and Roman letters typically denote statistics (x-bar for mean, s for standard deviation).

How big does a sample need to be?

Sample size depends on the research question, the type of analysis planned, the size of the effect the researcher expects to find, and the desired statistical power. For quantitative studies, run a power analysis before data collection. The power analysis takes the expected effect size, the desired statistical power (typically 0.80), and the alpha level (typically 0.05) and tells you the minimum sample size needed to detect the effect. For qualitative studies, sample size depends on saturation, the point at which additional participants stop revealing new themes, typically 12 to 30 participants for interview-based studies. For a step-by-step walkthrough including a worked G*Power example, see our sample size calculation guide.

What is the difference between probability and non-probability sampling?

In probability sampling, every member of the population has a known, non-zero chance of being included in the sample, and selection uses randomization. Probability sampling supports formal statistical inference to the population. In non-probability sampling, members of the population don't have known chances of selection, and the researcher uses other criteria (convenience, purpose, referral, or quota) to recruit participants. Non-probability sampling supports description of the sample and analytic generalization to similar contexts, but not formal statistical inference. The choice has direct consequences for the kinds of conclusions the study can defend.

Professional Editing for Your Research Manuscript

Reviewers screen the methodology section for population and sample definitions before they evaluate the analysis. A muddled or vague description of who the study is about and who actually provided data is one of the fastest paths to a desk rejection or a major-revisions decision. Clear writing in the methodology section is one of the strongest predictors of whether a manuscript moves forward in peer review.

Editor World provides dissertation editing and academic editing services for researchers preparing theses, dissertations, and journal article submissions. Every editor is a native English speaker from the United States, the United Kingdom, or Canada, with an advanced degree in their field. Every document is reviewed by a real person, never by AI. To see who would be working on your manuscript, you can choose your own editor from the Editor World roster, or request a free sample edit of up to 300 words before committing to a full edit. Pricing is fully transparent through an instant price calculator that shows your exact cost before you commit.

A certificate of editing confirming human-only native English editing is available as an optional add-on for journal submissions where AI use must be disclosed. For more on research methodology, see our research methodology guide, quantitative vs qualitative research guide, and research bias guide.

This article was reviewed by the Editor World editorial team. Editor World, founded in 2010 by Patti Fisher, PhD, provides professional editing and proofreading services for graduate students, academics, and researchers worldwide. BBB A+ accredited since 2010 with 5.0/5 Google Reviews and 5.0/5 Facebook Reviews. More than 100 million words edited for over 8,000 clients in 65+ countries.