Purposive Sampling Explained: What Is Purposive Sampling?

The purposive sampling method is a non-probability sampling technique where researchers make calculated choices in their sampling design to select a particular type of respondents. Also known as judgment sampling, subjective sampling, expert sampling, and selective sampling, it rests on the theory that sometimes researchers must pre-select subgroups from an entire population to create a case study or shape a grounded theory.

Purposive sampling is different from probability sampling techniques because it does not rely on random selection to find respondents. Even though random selection methods are the gold standard for scientific studies, purposive sampling can provide excellent qualitative research data, provided that it is linked to a coherent sampling strategy.

The following characteristics define purposive sampling.

1. 1. Non-probability sampling: Probability sampling studies select respondents using random chance. Purposive sampling has a strategy behind it, which eliminates the chance for true randomness.
2. 2. Relies on researchers’ judgment: Before they undertake a purposive sampling campaign, researchers make assumptions about a population and then intentionally reach out to selected members of that population to test their theory.
3. 3. More structured than convenience sampling: Convenience sampling is a form of non-probability sampling where researchers study respondents who are easy to recruit. Purposive sampling is also a form of non-probability sampling, but its lack of randomness owes to a thought-out research strategy that singles out certain people for careful study. Sometimes purposive sampling is less convenient than simple random sampling because it can be harder to reach targeted populations.

In real-world practice, purposive sampling takes on six primary forms.

1. 1. Typical case sampling: In this sampling design, a researcher intentionally looks for what they consider a representative sample of the population being studied. They intentionally reject any subjects whom they deem not representative of the population. Researchers and statisticians often use typical case sampling to examine a phenomenon of interest within the general population.
2. 2. Homogenous sampling: This purposive sampling method seeks saturation of a particular demographic or a group the researchers believe has shared characteristics. This research method has much in common with stratified sampling; however, stratified sampling starts by placing respondents into homogenous subgroups and then subjects those subjects to random selection. This makes it a form of probability sampling, whereas purposive sampling is, by definition, a non-probability sampling method.
3. 3. Maximum variation sampling: This research design is the opposite of homogenous sampling. It uses purposive selection to get the most heterogeneous (diverse) snapshot of the total population being sampled. This methodology lets researchers examine many possible outcomes within a single study group.
4. 4. Deviant case sampling: Also known as extreme case sampling, this research method intentionally looks for outliers in the overall sampling frame who are not representative of the population.
5. 5. Critical case sampling: In critical case sampling, subjects are selected based on researchers' inferences that they might represent a broader trend. Sometimes critical case sampling leads to the discovery of many more subjects who share the same traits with the respondents.
6. 6. Snowball sampling: In this method of sampling, respondents may be asked if they know other people who might qualify as study subjects. Respondents may be asked in person, or there may be an item on a questionnaire asking them for help in finding more people for the study. The recruited respondents can then, in turn, help recruit even more respondents, creating a snowball effect.

The use of purposive sampling comes with some clear advantages. Most prominently, this sampling approach clears out quantitative data that does not help researchers pursue a specific research objective. It lets scientists hone in on a particular subgroup of respondents—whether they are representative cases or deviant cases—and train their research questions on them without wasting time testing the entire population.

The drawback to purposive sampling is that it is only as good as a scientist's intuition. When a researcher opts for a purposive sampling method, they are making a bet that certain types of people are more worthy of study than others. Sometimes those bets pay off, but sometimes they do not. If a researcher's theory of the case was wrong, their purposive sampling may be rendered useless.

By contrast, truly random studies of populations produce results that are more consistently useful, regardless of a researcher's ability to generate strong hypotheses. As such, probability sampling methods (and their inherent randomness) remain the gold standard for the social sciences and the hard sciences, as well as polling.

Get the MasterClass Annual Membership for exclusive access to video lessons taught by science luminaries, including Terence Tao, Bill Nye, Neil deGrasse Tyson, Chris Hadfield, Jane Goodall, and more.