All research studies require information that is reliable. When it comes to writing a PhD thesis, writing a research paper, or even a survey, you need to make decisions on who you want to include.
In many cases it is not possible to study a whole population due to time, access and resources. Sampling methods are used to choose a subset and gather data that will yield information that is useful to the researcher's purposes.
The methodology impacts who gets represented and what it can conclude.
What Is Sampling in Research?
Sampling is the process of selecting persons, groups, objects or cases from a larger population. Researchers then choose the units, and analyze them to answer their research questions.
For instance, a researcher who is conducting a study on the experiences of PhD scholars who have submitted papers to academic journals, may not be able to reach all eligible scholars. The researcher will be able to choose a smaller sample size, and can gather data through questionnaires or interviews.
There are two general types of sampling methods:
Probability sampling | Non-probability sampling |
|---|---|
Uses random selection. | Uses non-random selection. |
All units have a nonzero probability of being selected and are known. | Probabilities of selection are generally not known. |
Can make statistical inferences with appropriate design. | Lacks the ability to generally apply simple statistics to the entire population most of the time. |
Equality of selection need not be maintained in the relationship. Select based on the research question and conclusions.
Population, Sample and Sampling Frame
Population is the entire set of items, persons or objects that a researcher is interested in. The target population may be PhD scholars who have submitted one or more manuscript to a scholarly journal for research with regards to academic publishing.
Sample is a part or subset of that population that is selected. The researcher may invite 100 eligible scholars to participate. This is a demonstrative sample size; the sample size to be used for the study should be explained for the specific study.
Sampling frame is a list or source from which potential participants can be identified (e.g. University register, professional directory, researcher database). Should include the target population as closely as possible. An outdated or incomplete frame may result in coverage issues and exclude eligible populations.
Identify the target population to which the study will relate from the accessible population that is within the reach of the researcher.
Research sampling is of two types of sampling in research: probability sampling and non-probability sampling. These contain a number of sampling techniques. A method of sampling that involves random selection is called probability sampling. All eligible units have a non-zero probability of being selected. It is useful when researchers want to make statistical inferences or estimate the characteristics of the population. 1. Simple Random Sampling All members of the sampling frame are equally likely to be selected. A random number generator or statistical software can be used to generate random numbers. In other words, a researcher who knows the entire set of potential PhD students can choose the number of students she needs for the sample-size plan of her study. The method involves the use of an appropriate sampling frame. 2. Systematic Sampling Researchers pick a random starting point and sample the units at regular intervals from an ordered list. For a list of 1000 researchers, and a planned sample of 100, the interval will be 10. The random start is followed by selection of every 10th person. Discuss the possibility of patterns developing in the list that might influence choices. 3. Stratified Sampling The population is divided into appropriate sub-groups (strata) and participants are randomly sampled from each strata. Scholars might be split by disciplines on a study of academic publishing and be chosen from each group. Stratification is useful for the inclusion of key subgroups. Selection may be by proportion to the size of the subgroups, or by different allocation when necessary. 4. Cluster Sampling The population is split into naturally occurring clusters or groups (in this case, universities or districts). A sample of clusters, selected at random, are sampled, or a sample is selected within each cluster and data gathered from all eligible units within each cluster. This will decrease fielding costs if the population is dispersed. 5. Multistage Sampling Multistage sampling is a way of sampling in which units are sampled in stages. A study may randomly select districts, universities within those districts and then researchers within the selected universities. In a probability-based design, a series of probability procedures are applied at each stage. It can be used in large studies with multiple levels of organisation. Non-probability sampling is a sampling method in which the participants are not selected by chance. Selection may be based on accessibility, participant characteristics, referral or a developing research need. These methods may be appropriate for exploratory and qualitative research, but they do not necessarily apply to inferences about a broader population. 1. Convenience Sampling The choice of participants is based on the availability and willingness to participate. A researcher may ask the scholars at a research workshop to fill in a questionnaire. It is practical but may be different from non-attendees who are researchers. 2. Purposive Sampling Researchers actively choose individuals who have experience or knowledge of what they are studying. Peer-review challenges research conducted by interviewing journal editors and researchers who have recently submitted manuscripts. The selection criteria should be clearly specified. 3. Quota Sampling Researchers establish a number goal for specific categories and recruit a number of participants until each number is met. A study could employ a specific number of PhD students, academics and research scientists. The selection within each quota does not have to be random as in stratified sampling. 4. Snowball Sampling First participants suggest other people who qualify for the study. This will help researchers find more groups that are hard to find on a typical list. For example, if a researcher is working in a specific niche, then they can also provide referrals to other researchers in the same niche. Some views might be overrepresented as recruitment is done through networks. 5. Theoretical Sampling It is a process that is primarily used in grounded theory, where additional cases and/or participants are selected as the data collection and analysis continue. If the problem in editorial communication is revealed after the initial interviews, the researcher can find some participants who can explain it. The selection is not just based upon the list, but following emerging concepts. Every research or study may require a different method. There is no single best method for every study. Select based on research question, population, design and intended conclusion. Get your research objective clear. If you are looking for estimates or statistical inferences of a defined population, then you should consider probability sampling. If you require detailed accounts from those who have a specific experience you may find purposive sampling more appropriate. Identify population and access. Establish inclusion/exclusion criteria. Determine if an accurate and complete sampling frame is available. Only units appearing on the frame can be selected at random. Match the design of the study with the method. The choice between probability sampling and non-probability sampling depends on the aims of the quantitative study. Purposive, snowball sampling, and theoretical sampling frequently are employed in qualitative studies. These are just some options available, and not rules. Take into account the time, cost and location. Cluster or multistage sampling may be practically useful for geographically scattered populations. Convenience sampling can be used for a pilot study but the limitations reported. Design the sample size by itself. The sampling method answers the question of who participates and the sample size answers the question of how many. When estimating population values, take into account confidence level, margin of error, population variability and sampling design. Effect size, significance level and statistical power are important for hypothesis testing. Some non-response should be expected as well. Unlike quantitative research, sample size in qualitative research is not typically determined using the type of statistical procedure. Researchers can employ either saturation or information power which takes into account the study aim, the specificity of the participants, theory, interview quality and analysis. Unlike quantitative research, sample size in qualitative research is not typically determined using the type of statistical procedure. Researchers can employ either saturation or information power which takes into account the study aim, the specificity of the participants, theory, interview quality and analysis. Coverage error may result from an incomplete sampling frame. Results may be skewed if those who respond are different from those who don't. Non-sampling errors are errors in measurement and data-processing. You should report in your methodology chapter: Target population Sampling frame Method of sampling Eligibility criteria Recruitment of the sample Your reason for the sample size Any limitations This provides readers with a judgment of how participants were chosen. Frequently Asked Questions What are the two main types of sampling methods? There are two types of sampling: Probability sampling and non-probability sampling. Probability sampling occurs when a sample is drawn at random with known selection probabilities, non-probability sampling is conducted when a sample is not drawn at random but rather through other means. What type of sampling techniques is used for qualitative research? Purposive sampling is widely used when the researcher requires participants that have experience in a field or topic. For specific population and designs, snowball or theoretical sampling might be appropriate. Which sampling method is suitable for quantitative research? Probability sampling is helpful when one wants to make statistical inferences about a defined population. This will depend on the population and the sampling frame. What is the major difference between stratified and cluster sampling? Stratified sampling is a sampling that is conducted among all the people in each defined stratum. Cluster sampling involves random selection of some groups and then either all or a sample of the units in the selected groups. Does the larger the sample, the better? No. Big sample size is no cure for selection or measurement error. The sample size should address the research design and analysis.
The choice of sampling method will depend on your research goal(s), population of interest, access to participants, and the analysis you will be using. Know how each method chooses participants, the limitations of the method and justify your choice in the methodology chapter. Looking for some assistance with the research process or a rationale for your methodology? Mikky Publication Services can assist researchers in selecting the type of research design, sampling, determining sample size, documenting the questionnaires, and planning the data collection and analysis. Learn more about our Research Methodology services to get assistance with creating a solid methodology that fits your research goals. 1. Acharya, A. S., Prakash, A., Saxena, P., & Nigam, A. (2013). Sampling: Why and how of it? Indian Journal of Medical Specialities, 4(2), 330–333. https://doi.org/10.7713/ijms.2013.0032 2. Cochran, W. G. (1977). Sampling techniques (3rd ed.). John Wiley & Sons. 3. Etikan, I., Musa, S. A., & Alkassim, R. S. (2016). Comparison of convenience sampling and purposive sampling. American Journal of Theoretical and Applied Statistics, 5(1), 1–4. https://doi.org/10.11648/j.ajtas.20160501.11 4. Lohr, S. L. (2022). Sampling: Design and analysis (3rd ed.). Chapman & Hall/CRC. 5. Malterud, K., Siersma, V. D., & Guassora, A. D. (2016). Sample size in qualitative interview studies: Guided by information power. Qualitative Health Research, 26(13), 1753–1760. https://doi.org/10.1177/1049732315617444Types of Sampling Methods in Research
A. Probability Sampling
Non-Probability Sampling
How to Choose the Right Sampling Method for Your Research
Sampling Error and Bias
Conclusion
References
Mikky Publication Experts
Mikky Publication Editorial
Published November 20, 2026
