Snowball and Respondent-Driven Sampling

Description

Recruiting hard-to-reach participants through referrals from existing participants, with RDS adding coupons and chain tracking to correct network bias.

Snowball sampling uses participants' social networks to reach populations with no sampling frame: specialists, stigmatised groups, informal workers or small professional communities. Each participant refers one or two others who meet criteria. Its weakness is homophily: chains reproduce the characteristics of the seeds, producing homogeneous samples. Heckathorn's respondent-driven sampling formalises the process for HIV research among hidden populations, issuing a limited number of coupons per participant, tracking recruitment chains, recording network size, and using these to weight estimates so that, with long enough chains, results become less dependent on seeds. In qualitative UX the main safeguard is monitoring diversity after early interviews and introducing new seeds from other parts of the network.

Attributes

Loss type
Siloed
Stage
Discovery
Effort
Weeks

Steps

  1. Recruit an initial seed participant who meets criteria and is known to others in the target population.
  2. After each interview, ask for 1–2 referrals. Get permission to use their name.
  3. RDS extension: Give each participant 2–3 formal recruitment coupons. Track chains. RDS attempts to correct network bias mathematically.
  4. Monitor for homophily: After 5–7 interviews, check whether the sample is inadvertently homogeneous. If so, introduce new seeds from different network parts.

Sources

  • Sampling and Choosing Cases in Qualitative Research
  • Qualitative Research from Start to Finish (Yin)
  • Snowball Sampling (Goodman)
  • Respondent-Driven Sampling: A New Approach to the Study of Hidden Populations (Heckathorn)
  • Gratton, C. & Jones, I., Research Methods for Sports Studies (2010)
  • Veal, A. J. & Darcy, S., Research Methods in Sport Studies and Sport Management (2014)

Also known as

  • Chain referral sampling
  • RDS
  • Network sampling
  • Referral recruitment

Categories

  • Research Operations
  • Research Design & Rigor