Snowball Sampling: What It Is and How To Use It in Research

By Indeed Editorial Team

Published August 4, 2021

Population research is useful for helping understand trends and experiences in a defined group of people. Certain research topics focus on specific populations that researchers might find difficult to find or contact. A method called snowball sampling can help researchers access those populations and understand them more successfully than other strategies might. In this article, we discuss snowball sampling, types of snowball sampling and some key advantages and challenges to consider prior to conducting research.

What is snowball sampling in research?

Snowball sampling, also known as chain-referral sampling, is a non-random sampling method used when test quantities are hard to find. In snowball sampling, people involved in your study nominate other potential sources you can use in your research. For example, it might be challenging to survey the top 1% of car owners who own premium luxury cars unless your sources share contact or referral information or provide your information to their contacts.

Related: 6 Sample Methods in Statistics (Plus Examples)

Types of snowball sampling

Different types of snowball sampling serve different purposes, and it can help to understand these types when deciding whether to use this strategy. Here are three snowball sampling types you might consider when conducting research:

Linear snowball sampling

This sampling technique helps researchers form a sample group one individual at a time. Each individual might refer an additional potential research subject, who then suggests another individual. This chain can continue until researchers achieve the ideal sample size for the topic they are investigating.

Exponential nondiscriminatory snowball sampling

In this type of sampling, researchers recruit the first subject of a sample population. That first subject can then provide multiple referrals for the study to recruiters. Each new referral can then also provide multiple referrals until there are enough subjects for the study.

Exponential discriminatory snowball sampling

This technique starts with the recruitment of one research subject. This subject can refer multiple potential subjects, and then researchers choose qualified candidates from these referrals. This can help researchers select the right subjects for their research.

Related: Types of Sampling: Choosing the Best Types and What To Avoid

Why is snowball sampling important?

Snowball sampling is helpful when research subject populations are unknown or difficult to reach. This can also help researchers gain a pool of applicable candidates to choose from in order to conduct their research. Snowball sampling is the preferred sampling method for the following research types:

  • Medical practices: Researchers can find individuals with less-researched diseases.

  • Social research: Researchers can gather as many participants as possible.

  • Cases of discord: Investigators can help identify witnesses of an event or instigators.

Related: Non-Probability Sampling: Definition and Types

Advantages of snowball sampling

There are several advantages to using snowball sampling to gather research subjects for your study. Here are some key advantages of using snowball sampling methods:

  • Quick sample sourcing: With subjects making referrals, researchers don't have to spend the time looking for subjects. This can help them focus their time and energy on conducting the research.

  • Cost-effective: Much of the sourcing is coming from sample referrals, which means less time taken to source a data population. This can lead to a reduction in cost for the study because researchers don't have to seek sources themselves.

  • Reach difficult target groups: Snowball sampling allows researchers to get access to difficult target groups because they are being referred to individuals within the target group from their primary source. Certain groups are hard to contact or even know outside of source referral.

  • Educate on population characteristics: These sampling methods can help researchers identify characteristics of the target population that they might not have known. This can help them make connections to the population and their research.

  • Minimal sourcing planning: With subjects making referrals, researchers don't have to make extensive sourcing plans. Many subjects may either come to them, the researchers, or subjects can name potential subjects for them to contact.

Related: A Guide To Statistics for Business

Disadvantages of snowball sampling

When using snowball sampling methods, it's important to consider some of the disadvantages that researchers might experience. Knowing potential obstacles prior to sourcing sample subjects can help you avoid these disadvantages and have successful research outcomes. Here are some key challenges to consider when using snowball sampling:

  • Sample bias: Due to the fact that sample subjects are referring to others in the population, there is a potential for sample bias to occur. Sample bias is when a sample population doesn't reflect the true distribution of the population and can skew the data one way or another.

  • Margin of error: With a potential sample bias occurring, that can leave a margin of error in researchers' data. This can mean that researchers miscalculate or interpret data.

  • Lack of cooperation: Certain data populations might be unwilling to come forward or share information in relation to research. This can lead to unsuccessful research outcomes or a lack of population data.

Tips to implement effective snowball sampling

There are several ways to counteract the potential obstacles of snowball sampling. The first way to overcome these obstacles is by being aware of them. Here are some other ways researchers can overcome potential obstacles:

  • Analyze responses as you receive them: Analyzing responses as you receive them can help you determine if your sample population has a bias towards certain responses. This can help you adjust your research method or help you determine if you need other sample subjects.

  • Establish your margin of error: Before starting the research, you should calculate your appropriate margins of error. This can help you identify skewed data and modify your research methods.

  • Encourage participation: Many participates might hesitate to share personal information. Encourage participation, rather than enforce it, to help participants feel comfortable. Additionally, anonymous participation might help them feel more comfortable sharing their responses.

How to use snowball sampling in research

Although specific steps can vary depending on your research subject or sampling method, many snowball sampling techniques follow a similar general pattern. Here is how you can apply snowball sampling to your research:

1. Identify potential population subjects

The first step in effectively using the snowball sampling method is to identify a sample subject. This step may likely result in one or two sources to begin with. It's important to verify that these potential subjects fill the subjective requirements of your research, such as age or occupation.

2. Contact potential subjects

Once you have identified your sample subject or subjects, you can contact them and determine if they are interested in participating in your research. There are many ways to communicate with subjects, such as in-person, by phone or email communications. If subjects wish to participate, you can continue with your sampling steps. If they don't want to be a part of your research, you can either contact your second sample source or find other sources.

3. Ask to join in research

When asking participants to join your research, educate them on the research and what you hope to achieve. This can help participants make sound decisions on whether they wish to participate. Even if subjects don't wish to participate, you can still inquire about subject referrals.

4. Encourage subject referral

Once you have the initial subjects enrolled in your research, you can then encourage them to offer subject referrals. You might encourage participants to recommend that other participants come forward rather than directly naming other potential subjects. This can help prevent any embarrassment for referred subjects when you are researching delicate topics.

5. Evaluate referrals if using discriminative sampling

If you are using discriminative sampling methods, once you get referral information you can analyze it to determine the most applicable participants. This can help identify subjects that fit your topic's research objectives, such as subjects fitting a specific age or occupation criteria.

6. Repeat until you reach the desired sample size

Once you have contacted your initial referrals, you can then ask them for additional referral information. You may repeat these steps until you reach the desired population or until you have exhausted all referrals. You might also stop gathering participants when you determine that you have accumulated enough of a population to provide useful data.

Explore more articles