Snowball sampling
Snowball sampling is a non-probability sampling method in which existing participants recruit or refer further participants from among people they know, so the sample grows like a rolling snowball. It is used to reach populations that are hard to find through lists.
How it works
The researcher starts with a few people who meet the criteria, the seeds, and asks each to suggest others who do. Those referrals are screened, interviewed and asked for further referrals, until the study has enough participants or the chains run dry. Starting with several seeds from different parts of the population widens the eventual sample.
The method suits hidden, rare or specialized groups: people with a particular condition, members of a profession with no public directory, or users of a niche product. Because people refer others like themselves, the sample tends to cluster, so findings describe the networks reached and not the whole population.
A worked example
A medical device company needs to interview 15 surgeons who have used a specific robotic system. It starts with 4 seeds from its own customer list, each of whom refers 2 or 3 colleagues. Within 3 weeks the study has 17 surgeons from 11 hospitals, 9 of them outside the company's own accounts.
Related
Automate this in Pulse
For a hard-to-reach group a team might otherwise snowball, Pulse can recruit directly to a profile in the brief, avoiding samples clustered in one referral network.