When students write their methodology chapter, they often use the terms population, sampling frame and sample as if they mean the same thing. They do not. Mixing them up can confuse examiners and, more importantly, can hide problems with how participants were selected. Understanding the difference helps you design a better study, describe it accurately and judge how far your findings can be generalised.
This post explains each term, shows how they relate to each other, gives practical examples, and highlights common errors to avoid.
The big picture
Think of the three terms as moving from the broad group you want to study to the specific people you actually study:
- Target population: the entire group you want your findings to apply to.
- Sampling frame: the actual list or source from which you can select participants.
- Sample: the participants you actually select and collect data from.
In an ideal study, the sampling frame matches the population perfectly, and the sample is a good representation of the sampling frame. In practice, there are usually gaps at each step, and good research reports them honestly.
What is a population?
The population is the complete set of individuals, organisations, events or items that share the characteristics you are interested in. It is defined by your research question.
Researchers sometimes distinguish between two levels:
- Target population: the full group you would ideally like to generalise to, for example, all nurses working in public hospitals in Malaysia.
- Accessible population: the part of the target population you can realistically reach, for example, nurses working in public hospitals in two states that have granted you access.
A clear population definition usually includes who, where and when, along with any inclusion and exclusion criteria. For example: "Full-time registered nurses with at least one year of experience, working in public hospitals in Selangor and Johor during 2026."
What is a sampling frame?
The sampling frame is the concrete list, database, register or procedure used to identify and select members of the population. It is the bridge between the population you have defined and the sample you draw.
Examples of sampling frames include:
- A university's list of enrolled students.
- A professional body's membership register.
- A company's employee database.
- A government business directory.
- A list of schools obtained from the education ministry.
- Customer records or loyalty programme databases.
When no list exists, the sampling frame may be a procedure rather than a document. For example, intercepting every fifth customer leaving a store during a specified period defines who is eligible to be sampled.
What is a sample?
The sample is the subset of the sampling frame that you actually select and from whom you collect data. It is important to distinguish between:
- Intended or selected sample: the people you invite to participate.
- Achieved sample: the people who actually respond and provide usable data.
The difference between the two is your response rate. A low response rate can introduce non-response bias if those who respond differ systematically from those who do not.
A worked example
Suppose your research question is: "What factors influence job satisfaction among secondary school teachers in Kuala Lumpur?"
- Target population: all secondary school teachers in Kuala Lumpur.
- Sampling frame: a list of government secondary schools in Kuala Lumpur, and the staff lists provided by the schools that agreed to participate.
- Selected sample: 400 teachers chosen using stratified random sampling from those staff lists.
- Achieved sample: 312 teachers who completed the questionnaire.
Notice that private school teachers are not in the sampling frame, so your findings may not apply to them. This should be acknowledged as a limitation or reflected in a narrower population definition.
Why the sampling frame matters so much
The quality of your sample cannot be better than the quality of your sampling frame. Common sampling frame problems include:
- Undercoverage: some members of the population are missing from the list, such as new employees not yet added to a database.
- Overcoverage: the list includes people who are not part of the population, such as former employees who have left.
- Duplicates: some individuals appear more than once, giving them a higher chance of selection.
- Outdated information: contact details are wrong or incomplete.
- Clustering: one entry represents several people, such as a household address.
These issues create coverage error, which is the difference between the population and the sampling frame. Coverage error can bias your results even if you use a perfectly random sampling method.
Sampling frames in qualitative research
Qualitative studies usually use non-probability sampling, such as purposive or snowball sampling, and do not aim for statistical generalisation. However, the concepts are still useful. You should still define the group you are interested in, describe how you identified potential participants, for example through a professional network, gatekeepers or online communities, and explain who took part. This transparency helps readers judge the transferability of your findings.
How to describe them in your methodology chapter
- Define your target population precisely, including inclusion and exclusion criteria.
- State your accessible population if it differs from the target population.
- Describe your sampling frame, its source, how up to date it is and any known limitations.
- Explain your sampling technique and how you calculated or justified your sample size.
- Report the number invited, the number who responded, and the final number used in analysis.
- Discuss how differences between population, frame and sample may affect generalisability.
Common mistakes to avoid
- Using "population" to mean the people who answered your survey.
- Not mentioning the sampling frame at all.
- Claiming findings apply to a population that the sampling frame did not cover.
- Ignoring the gap between the selected and achieved sample.
- Describing convenience sampling as random sampling.
Final thoughts
The population is who you want to understand, the sampling frame is how you can reach them, and the sample is who you actually study. Keeping these three ideas separate helps you design a sound sampling strategy, report your methods clearly and make honest claims about generalisability. Examiners and reviewers notice when these terms are used precisely, and it strengthens the credibility of your whole study.