Non-probability sampling

Non-probability sampling represents a group of sampling techniques that help researchers to select units from a population that they are interested in studying. Collectively, these units form the sample that the researcher studies [see our article, Sampling: The basics, to learn more about terms such as unit, sample and population]. A core characteristic of non-probability sampling techniques is that samples are selected based on the subjective judgement of the researcher, rather than random selection (i.e., probabilistic methods), which is the cornerstone of probability sampling techniques. Whilst some researchers may view non-probability sampling techniques as inferior to probability sampling techniques, there are strong theoretical and practical reasons for their use. This article discusses the principles of non-probability sampling and briefly sets out the types of non-probability sampling technique discussed in detail in other articles within this site. The article is divided into two sections: principles of non-probability sampling and types of non-probability sampling:

Principles of non-probability sampling

There are theoretical and practical reasons for using non-probability sampling. In addition, you need to decide whether non-probability sampling is appropriate based on the research strategy you have chosen to guide your dissertation.

Theoretical reasons

Non-probability sampling represents a valuable group of sampling techniques that can be used in research that follows qualitative, mixed methods, and even quantitative research designs.

Despite this, for researchers following a quantitative research design, non-probability sampling techniques can often be viewed as an inferior alternative to probability sampling techniques. Non-probability sampling techniques can often be viewed in such a way because units are not selected for inclusion in a sample based on random selection, unlike probability sampling techniques. As a result, researchers following a quantitative research design often feel that they are forced to use non-probability sampling techniques because of some inability to use probability sampling (e.g., the lack of access to a list of the population being studied). However, this is not the case for researchers following a qualitative research design.

When following a qualitative research design, non-probability sampling techniques, such as purposive sampling, can provide researchers with strong theoretical reasons for their choice of units (or cases) to be included in their sample. Rather than using probabilistic methods (i.e., random selection) to generate a sample, non-probability sampling requires researchers to use their subjective judgements, drawing on theory (i.e., the academic literature) and practice (i.e., the experience of the researcher and the evolutionary nature of the research process). Unlike probability sampling, the goal is not to achieve objectivity in the selection of samples, or necessarily attempt to make generalisations (i.e., statistical inferences) from the sample being studied to the wider population of interest. Instead, researchers following a qualitative research design tend to be interested in the intricacies of the sample being studied. Whilst making generalisations from the sample to the population under study may be desirable, it is more often a secondary consideration. Even whether this is desired, there are additional problems of bias and transferability (or validity) [see the section on Research Quality for more information on research strategies, sampling techniques, and bias].

Practical reasons

Non-probability sampling is often used because the procedures used to select units for inclusion in a sample are much easier, quicker and cheaper when compared with probability sampling. This is especially the case for convenience sampling. For students doing dissertations at the undergraduate and master's level, such practicalities often lead to the use of non-probability sampling techniques.

As mentioned, for researchers following a quantitative research design, non-probability sampling techniques can often be viewed as an inferior alternative to probability sampling techniques. However, where it is not possible to use probability sampling, non-probability sampling at least provides a viable alternative that can be used. As such, it ensures that research following a quantitative research design is not simply abandoned because (a) it cannot meet the criteria of probability sampling and/or (b) meeting such criteria is excessively costly or time consuming, such that it would not be sponsored. This could significantly diminish the potential for researchers to study certain types of population, such as those populations that are hidden or hard-to-reach (e.g., drug addicts, prostitutes), where a list of the population simply does not exist. Here, snowball sampling, a type of non-probability sampling technique, provides a solution.

Non-probability sampling can also be particularly useful in exploratory research where the aim is to find out if a problem or issue even exists in a quick and inexpensive way. After all, you may have a theory that such a problem or issue exists, but there is limited or no research that currently supports such a theory. Where your main desire is to find out is if such a problem or issue even exists, the potential sampling bias of certain non-probability sampling techniques can be used as a tool to help you. For example, you may choose to select only those units to be included in your sample that you feel will exhibit the problem or issue you are interested in finding. If this problem or issue does not exist even in your biased sample, it is unlikely to be present if you selected a relatively unbiased sample (whether using another non-probability sampling technique; or even a probability sampling technique). This would help you to avoid a potentially more time consuming and expensive piece of research looking into a potential problem or issue that actually doesn't exist. It may also be considered an ethical approach to finding out whether a problem or issue is worth examining in more depth, since fewer participants are subjected to a research project unnecessarily.

Deciding whether non-probability sampling is appropriate

If you are considering whether to use non-probability sampling, it is important to consider how your choice of research strategy will influence whether this is an appropriate decision. Even if you know that non-probability sampling fits with the research strategy guiding your dissertation, it is important to choose the appropriate type of non-probability sampling techniques. These non-probability sampling techniques are briefly set out in the next section.

Types of non-probability sampling

There are five types of non-probability sampling technique that you may use when doing a dissertation at the undergraduate and master's level: quota sampling, convenience sampling, purposive sampling, self-selection sampling and snowball sampling.

To get a sense of what these five types of non-probability sampling technique are, imagine that a researcher wants to understand more about the career goals of students at a single university. Let's say that the university has roughly 10,000 students. These 10,000 students are our population (N). Each of the 10,000 students is known as a unit (although sometimes other terms are used to describe a unit; see Sampling: The basics). In order to select a sample (n) of students from this population of 10,000 students, we could choose to use quota sampling, convenience sampling, purposive sampling, self-selection sampling and snowball sampling: