Non-Probability Sampling Methods

There are also multiple types of non-probability sampling that you can pursue in your own research. Here, we will go over convenience sampling, quota sampling, snowball sampling, and site-based sampling.

Convenience sampling is when a researcher locates nearby participants that meet their inclusion criteria. These participants are chosen because of their convenience in terms of both proximity and willingness to participate in the study (Robinson, 2014). 

For example, if a psychology professor asks their students to take part in a study, these students are a sample of convenience because they are easy to access and will most likely agree to participate. Convenience sampling is often considered to be the easiest, but not the strongest form of sampling (Roulston, 2018). 

Participants selected using convenience sampling tend to have similar characteristics because they are sampled from the same space. If we go back to our example of the psychology professor asking their students to participate in a research study, these participants would all share the characteristic of being a student at a particular college. 

Quota sampling is when the researcher identifies specific categories, defined by certain characteristics, and seeks to fill these categories with a predetermined number of people. Accordingly, all the key characteristics that the researcher is interested in are represented in the sample (Robinson, 2014). When using quota sampling, the sample size tends to be more strict. because you are looking for a specific number of participants in each category. The categories that you choose should focus on the people who are most likely to provide insight into your research topic (Mack et al., 2005). 

For example, say you are conducting a study about age and parenthood among women. You might look to fill categories based on age and gender with a certain number of female participants who are 20-25, a certain number who are 26-31, and a certain number who are 32-37.

Snowball sampling, also known as chain-referral sampling, is when a researcher locates a few initial participants and then relies on these participants to refer them to others (Evans & Rooney, 2011). Imagine a snowball rolling down a hill. As the snowball goes further, it will pick up more snow and become larger. In the same way, as you continue to get referrals, your sample size will increase over time. However, snowball sampling is highly related with what is known as “self-selection bias”  which occurs when participants choose whether or not to take part in a study (Ellard-Gray et al., 2015). The group that chooses to take part is probably not characteristically equivalent to the group that chooses not to take part, and as a result, an overarching insight into the population of interest is not achieved (Glen, 2017). 

Snowball sampling is especially useful when recruiting from a hard-to-reach population, and sometimes may be your only option in this case (Ellard-Gray et al., 2015). It allows you to access a population through one or two individuals who act as points of entry into the larger group. One drawback to snowball sampling is that, depending on your chosen point of entry, you could end up with a biased sample.

Site-based sampling is designed to create a truly representative sample centered around the places, organizations, and/or services used by your population of interest. The first step is to specify inclusion and exclusion criteria. Once you set the boundaries of your sample, you can then create a list of the places, organizations, and services used by your population of interest. A site-based approach ensures that your sample will have the necessary characteristics that you are interested in studying, while helping you to build a relationship with the community you will be working in (Arcury & Quandt, 1999).