Snowball Sampling: Exploring What It Is, Its Uses And Benefits

In today’s increasingly complex world, there are ever-growing ways of conducting research, as individuals and businesses look to develop new theories, validate existing ones or solve specific problems. And while all this research broadly falls in to one of two camps namely quantitative or qualitative, along with their respective approaches to sampling that includes probability and non-probability; the choice of sampling methods continues to grow.

Today we will explore a non-probably sampling method called snowball sampling, including some of its most common applications and the benefits for those who choose to use it. But before we do that, we’ll provide a brief introduction to sampling and the non-probability sampling method.

What is sampling?

When we think about sampling in research, it’s a bit like randomly selecting names from a phone book or drawing lots to select a winner for a competition. It’s about picking a smaller part of a much larger group, and in doing so making that research more achievable. This subset, which represents the entire population of interest is referred to as a sample.

So, why is this so important?

Well, just imagine trying to survey a few million people for your research, which is totally impractical.

This is why sampling is vital. By examining a smaller group, researchers can learn important things without overwhelming themselves, and make more accurate conclusions as a result.

Types of sampling

As we’ve already touched on, there are two main types of sampling in research: probability and non-probability.

• Probability sampling: ensures everyone in a group has an equal chance of being chosen
• Non-probability sampling: with this more subjective approach, there is more focus on specific criteria and personal judgement

Focusing on non-probability sampling

The non-probability sampling approach is based on a method where not everyone in a population has an equal chance of being selected. In contrast to probability sampling, it relies more on specific criteria and personal judgment, but this can be useful in situations where there is a lack of time and limited resources.

There are several methods under non-probability sampling. Here are some of the most common ones:

• Convenience sampling: the basis of convenience sampling is built on selecting people who are easiest to reach, which could be those within closest proximity or those who are currently available
• Judgmental or purpose sampling: here participant selection is based on the researcher’s judgment, such as experts or those with specific knowledge
• Quota sampling: under this approach the population is divided into groups with participants selected from each group, until a set quota is met
• Snowball sampling: here a single participant is initially selected to grow the sample by referring other participants in a chain-like effect

What is snowball sampling?

Snowball sampling is a research method that is typically used to study a hard to reach population.

Think about if you were trying to gather snowflakes in a snowball. Imagine starting with one small snowball and rolling it down a hill. On the way down, you’ll pick up more and more snowflakes.

Well, snowball sampling is very much like that, but in research you’ll get the methodological snowball effect, where your sample size gradually expands.

Types of snowball sampling

When it comes to snowball sampling, there’s a lot more to it than initially meets the eye. This is reinforced by the fact that it is also commonly referred to as ‘chain’, ‘respondent-driven’, ‘network’ and ‘seeded’ sampling – each with their own characteristics and perspectives.

Let’s take a look at them.

Chain referral sampling

With this approach, participants refer to others sequentially. This results in a chain-like structure.

• It’s easy and straightforward, which is why it’s comparatively simple to implement
• It can be effective when you’re trying to study communities with a clear social structure, by actively encouraging participants to take part in the referral process
• It can be used in instances where network building is a key aim
• It’s important to remember that it’s only suitable for participants who are comfortable with referring others

Respondent-driven sampling

This method combines elements of snowball and probability sampling

• It improves data accuracy, by providing participants with information about their social networks
• Sampling biases are corrected through statistical techniques, which helps improve sample representativeness and enhance reliability by reducing selection bias
• It enables a systematic method for determining participant incentives
• It helps create a sense of trust in the participants

Network sampling

As the name indicates, network sampling is based on a method whose foundation relies on building connections.

• It focuses on whole social networks, with an emphasis on studying interactions and relationships
• Providing a holistic view of community structures and social dynamics, network sampling helps give you a bird’s eye view of the top influencers in your community
• It also provides a snapshot of the interconnectedness of a study’s participants
• This approach is often used in sociology and anthropology studies to analyse social relationships

Seeded snow sampling

Before looking at this one in a bit more detail, it’s best to explain what a seed is in snowball sampling.

Well, a seed is like the initial domino in a row. This is because snowball sampling begins with one or a single set of participants, which is like the initial seed that starts the chain.

Seeded snowball sampling enables you to control the initial participants, which helps maintain specific characteristics or expertise.

• This is the ideal approach if you wish to focus on specific subgroups within a larger community
• It helps provide a solid and structured approach to building a participant network
• Given how important it is to get off to a good start, seeded snowball sampling ensures a consistent starting point or a sampling process
• Seeded snowball sampling can help balance inclusivity and specificity in participant selection

The user base for snowball sampling

At this point, it can be helpful to know who uses snowball sampling.

The great thing about snowball sampling is that it’s quite adaptable in nature. This ensures it has quite a diverse user base. Let's take a look at a few of these.

Human rights organisations

Given that it’s an effective way of reaching out to persecuted groups, snowball sampling is a popular method for human rights organisations.

It helps them to gather first hand narratives of human rights violations. This enables them to advocate for the right cause and strengthens their policy influence by providing concrete data. Besides that, it also supports legal actions.

Snowball sampling helps educate the public about community challenges, while the data it helps to collect empowers affected communities through giving them a platform to share their stories.

Moreover, it helps support the efforts of long-term change by exposing recurring human rights issues.

Public health research

The snowballing approach also helps aid the understanding of disease transmission behaviours, which can help the growth of targeted development strategies.

By highlighting areas requiring intervention and investment, it can help reveal any disparities in healthcare access.

Consequently, snowball sampling is effective in supporting epidemiological studies, whose data collection can help aid future health education initiatives. In addition, it helps provide effective data for vaccine research and is beneficial in the study of less-known diseases.

Market research

When it comes to a company’s growth, market research is one of the most effective areas to help them. This is because snowball sampling offers an insight into consumer behaviour, as well as helping to identify distinct consumer segments for targeted marketing. This can also help you to make better predictions about market trends and more informed decisions as a result.

Social science

Given its ability to break down social dynamics within communities, snowball sampling can be very effective in social science and the study of social movements, identities and community development efforts.

Snowball sampling can also be hugely effective in supporting identity studies, and can help you delve into the depths of gender, and social identity within communities.

Benefits of snowball sampling

If you’ve read this far, you’ve probably noticed many of the advantages of snowball sampling already. However, it can be handy to have a list of the key benefits in one place, which we've outlined below.

• Easier access: the great thing about this method, is that it helps researchers reach more people, especially those who are less easily accessible using more traditional research methods
• Greater trust and comfort: participants feel much more comfortable talking if they’re introduced to someone they know, which also helps build greater trust in the overall research process
• More diverse perspectives: it can help gather a wider pool of diverse opinions and experiences, further enhancing your overall understanding of your topic of interest
• More cost-effective: compared to many other methods snowball sampling is budget-friendly, as it relies on existing social networks
• Greater flexibility: it adapts well to different communities and situations, which also makes this approach a good fit for a wide variety of research needs
• Ideal for investigating sensitive topics: with this approach participants feel more comfortable to discuss sensitive issues in a familiar and trusted environment

Limitations of snowball sampling

As with any method, there are always disadvantages as well as benefits. So, it’s always best to be fully aware of these, so you can make a more informed decision about the approach that’s right for you.

• More limited selection: snowball sampling may inadvertently leave out some important individuals
• It's not random: your research can be subject to potential bias, as people are not selected randomly
• Inaccurate results: your results may not accurately represent the whole community
• Privacy concerns: some participants may not share their personal information openly
• Time-consuming: it can take a lot of time to locate and then interview participants
• Biased data: any initial biases can quickly grow and potentially harm the overall quality of your research

Wrapping up

We hope you found this blog interesting and useful. And if you weren’t already familiar with snowball sampling, you feel more informed about its benefits and where you can use it.

The key thing about snowball sampling, is that it helps give you access to and opinion from groups that would otherwise be really hard to reach. So, even though you have to be careful about any biases and potential inaccuracies that can arise, it’s an approach that’s well worth exploring for the wealth of insights it can provide.