Snowball sampling: What it is and when to use it

Ever struggled to find niche demographics for a research project? In these situations, snowball sampling can help. 

The idea behind snowball sampling is that it’s a way to grow your pool of participants from a small start. This lets you tap into hard-to-reach groups and uncover insights that might otherwise stay hidden.

In this guide, you'll learn what snowball sampling is, when to use it, and how it can benefit your research. Whether you're an academic studying rare phenomena or a market researcher trying to understand niche consumer groups, snowball sampling might be what you need.

What is snowball sampling?

Let’s set the scene: you're at the top of a snowy hill with a tiny snowball in your hand. As you roll it down the slope, it picks up more and more snow, getting bigger and bigger. That's snowball sampling in a nutshell. But instead of snow, you're gathering research participants.

Here's how it usually goes:

1. You start with a small group of participants who fit your study criteria.
2. These initial participants then point you to others in their network who might be a good fit.
3. Those new recruits, in turn, refer even more potential participants.
4. The process keeps going, with your sample size "snowballing" until you hit your target number or run out of new leads.

The good thing about snowball sampling is how it taps into social networks and word-of-mouth to reach people who might be tricky to find through other methods. It's particularly handy when you're looking at marginalized groups, rare conditions, or sensitive subjects where people might not be keen to come forward on their own.

When does snowball sampling work best?

Snowball sampling isn't a magic bullet for every research project. It's more like a specialized resource in your research toolkit. Here are some situations where it really shows its worth:

Hidden or hard-to-reach groups

Trying to study groups that are stigmatized, involved in illegal activities, or just plain hard to find? Snowball sampling can help you make those first key connections because it taps into existing social networks. 

People who are part of these hidden groups often know others like them, creating a chain of referrals that can lead you to participants you might never have found otherwise.

Rare conditions or experiences

If you're looking into uncommon medical conditions, unique life experiences, or niche interests, snowball sampling can help you find enough people to draw meaningful conclusions. Using personal networks makes it easier to locate individuals with rare attributes that are otherwise difficult to identify.

Exploratory research

When you're trying to get a handle on a phenomenon or come up with new ideas in the early stages of a project, snowball sampling can help you find unexpected insights and directions for further study. You can adapt and refine your research focus, based on the new information gathered from initial participants.

Sensitive topics

For research dealing with personal or controversial subjects, people might feel more comfortable joining if someone they trust referred them.

Limited resources

When you're short on time or money for more extensive sampling methods, snowball sampling can be a cost-effective way to build your participant pool. With participants recruiting more participants, you don’t have to rely on extensive outreach efforts, saving valuable resources.

The aim is to match your sampling method to your research goals and the nature of your target population. If you want a truly representative sample, you might need to try other techniques or approaches - you can learn more about these in our complete guide to representative samples.

Different types of snowball sampling

There are different types of snowball sampling, each with its own strengths:

Linear snowball sampling

This is the simplest form, where each participant refers just one other person. It's like a chain, with each link connecting to the next. This method works well when you want to keep tight control over the sample growth or when you're dealing with a very specific or small population.

Exponential discriminative snowball sampling

Here, participants can refer multiple others, but you pick and choose which referrals to include based on specific criteria. This approach lets you grow your sample faster while still keeping some control over the direction of the study.

Exponential non-discriminative snowball sampling

In this version, you include all referred participants who meet your basic criteria. It's the fastest way to grow your sample size, but it can also mean less control over who ends up in your study.

Your choice between these methods depends on: 

• What you need for your research
• How much time you have
• Who you're trying to study

For instance, if you're looking into a rare medical condition, you might go for exponential non-discriminative sampling to cast the widest net possible. On the other hand, if you're exploring a sensitive social issue, linear sampling might help you maintain trust and privacy throughout the referral chain.

Snowball sampling in action

Let's look at some some real-world examples of snowball sampling:

HIV/AIDS research

In the early days of the HIV/AIDS epidemic, researchers used snowball sampling to reach affected individuals who often didn't want to come forward, or were hard to find. By starting with a few willing participants and tapping into their networks, scientists were able to gather key data about how the disease was spreading and affecting people.

Undocumented immigrant studies

Researchers looking into the experiences of undocumented immigrants often use snowball sampling to build trust and access this hidden population. Initial contacts within immigrant communities can refer others who might be willing to share their stories, giving valuable insights into their challenges and needs.

Rare disease research

When studying extremely rare medical conditions, finding enough participants through traditional methods can be highly challenging. Snowball sampling lets researchers tap into patient networks and support groups, connecting with people who have the condition and their families.

Subculture studies

Anthropologists and sociologists studying specific subcultures—like underground music scenes or niche hobby groups—often use snowball sampling to get their foot in the door and build rapport. A few key contacts can open up the broader community, allowing for rich, in-depth observations.

Business network analysis

In the corporate sector, snowball sampling can map out informal networks within organizations. Starting with a few key employees, researchers have the ability to uncover how information and influence flow through a company, revealing insights that might not show up on the official organisation chart.

These examples show how versatile snowball sampling can be across different fields and research contexts. Tapping into existing social networks means researchers can access valuable data that might otherwise stay out of reach.

The pros and cons of snowball sampling

Like any research method, snowball sampling has its pros and cons. Let's break them down:

Wrapping up: Is snowball sampling right for you?

Snowball sampling can be  powerful, especially when traditional sampling methods fall flat. By tapping into social networks and word-of-mouth referrals, it has the potential to open doors to populations and insights that might otherwise stay hidden. While it's not perfect, knowing when and how to use snowball sampling can give your research a real boost, especially when you're exploring new territory or trying to reach elusive groups.

So, next time you're faced with a tricky research population, consider giving snowball sampling a try. You might be surprised at how quickly your sample—and your insights—start to snowball.

Find out more about sampling methods for online research in our complete guide to representative samples.