Snowball sampling

How Snowball Sampling Works

The application of snowball sampling generally involves several stages.

Stage 1: Identify Initial Participants. Researchers begin by recruiting one or more participants who possess characteristics relevant to the study. This is often the most challenging stage because suitable participants can be difficult to locate.

Stage 2: Request Referrals. Researchers ask the initial participants to identify additional individuals who meet the study requirements.

Stage 3: Recruit New Participants. The referred participants are contacted and invited to participate.

Stage 4: Continue the Referral Process. Each new participant may then identify further potential participants. The sample gradually expands through referral chains.

Stage 5: Stop Recruitment. Recruitment usually stops when:

• the desired sample size is reached
• data saturation occurs
• no new referrals emerge
• additional recruitment becomes impractical

Snowball Sampling in Business Research

Snowball sampling is widely used in business and management research when studying specialised professional groups. For example:

• Researchers investigating startup founders may use referrals among entrepreneurial networks.
• Studies involving venture capital investors often rely on industry contacts.
• Research examining senior executives may recruit participants through professional introductions.
• Studies of management consultants may use referrals within consulting firms.

Suppose a researcher wishes to examine leadership styles among senior managers at McKinsey & Company. Obtaining direct access to a large number of senior consultants may be difficult. However, after interviewing one senior manager, that participant may introduce the researcher to colleagues who also satisfy the study requirements.

Similarly, studies involving employees within a specific organisation often use snowball sampling because existing participants can help identify additional colleagues willing to participate.

Common Mistakes

One misunderstanding frequently encountered among students is assuming that snowball sampling automatically produces representative samples. Because participants recruit others from their own networks, certain viewpoints may become overrepresented. Researchers sometimes fail to recognise the influence of network bias. Individuals often refer people who share similar backgrounds, experiences, or perspectives, reducing diversity within the sample.

Another challenge occurs when researchers rely excessively on a single referral chain. This may produce a sample dominated by one particular subgroup while excluding others. Some students also neglect ethical considerations associated with referrals. Participants should never feel pressured to disclose personal contact information without consent.

Finally, snowball sampling is occasionally selected simply because participant recruitment appears easier, even when more appropriate sampling methods are available.

Advantages and Disadvantages of Snowball Sampling

One major advantage of snowball sampling is its ability to access hidden or difficult-to-reach populations. Researchers can often recruit participants who would otherwise remain inaccessible.

Another strength is cost-effectiveness. Existing participants help identify new participants, reducing recruitment effort and expense. The method is also relatively quick to implement and often requires limited administrative planning at the beginning of the study. Snowball sampling can be particularly valuable when trust plays an important role in participant recruitment because referrals from known contacts may increase participation rates.

However, snowball sampling also has important limitations. Sampling bias represents one of the most significant weaknesses because participants tend to recruit individuals from similar social or professional networks. As a result, findings may not represent the wider population accurately.

Another limitation is the inability to calculate sampling error or make strong statistical generalisations because participant selection is non-random. Researchers may also encounter ethical concerns when participants are asked to provide details of friends, colleagues, or professional contacts. Finally, the success of the sampling process depends heavily on participants’ willingness to provide referrals.

Snowball Sampling in the Age of AI and Digital Research

Digital technologies have significantly expanded opportunities for snowball sampling. Professional networking platforms such as LinkedIn, industry forums, online communities, startup ecosystems, and specialist social media groups make it easier than ever for researchers to identify and recruit participants through referral networks. Digital communication tools also allow referral chains to develop rapidly across geographical boundaries.

Researchers studying cryptocurrency investors, technology entrepreneurs, remote workers, AI specialists, or digital content creators frequently use online referral networks to recruit participants. AI-powered networking platforms can also help identify connections between individuals, organisations, and professional communities that may support participant recruitment.

At the same time, digital environments introduce new challenges. Online referral chains may become concentrated within algorithmically connected communities, creating echo chambers that reduce sample diversity. Researchers must also consider issues relating to privacy, informed consent, authenticity of online identities, and potential misuse of personal information.

Although digital technologies have made snowball sampling more efficient, researchers must continue to evaluate sample diversity carefully and acknowledge limitations relating to representativeness and network bias.

Need help justifying snowball sampling in your dissertation methodology chapter?

The Dudovskiy AI Research Assistant can explain when snowball sampling is appropriate, identify its limitations, and generate a dissertation-ready justification linked to your research objectives.

When to Use Snowball Sampling

Snowball sampling is most appropriate when:

• the target population is difficult to identify
• no complete sampling frame exists
• participants possess specialised knowledge or experience
• access depends on trust and personal connections
• the population is relatively small or hidden
• exploratory or qualitative research is being conducted

For example, snowball sampling may be suitable when studying startup founders, venture capital investors, senior executives, expatriate managers, luxury club members, cybersecurity specialists, or employees within restricted organisational environments.

Use snowball sampling when participants are difficult to locate directly and existing participants can help identify others who meet the study requirements.

Exam Tip

Many students justify snowball sampling simply by stating that participants are “difficult to find.” This justification is often incomplete. Examiners usually expect you to explain why a sampling frame is unavailable, why alternative sampling methods are impractical, and how referral-based recruitment helps achieve the research objectives. A strong justification also acknowledges limitations relating to representativeness and network bias.