Snowball sampling is often used in qualitative research when a population is hard-to-reach or hidden. This makes it useful for sensitive topics too.
This sampling technique is called “snowball” for this process. Like a snowball rolling downhill, the sample grows larger. Each new referral adds more people to the group.
Non-probability sampling means that researchers, or other participants, choose the sample rather than selecting it at random. Not every population member has an equal chance of being included in the study.
Probability sampling is the opposite approach. It gives every member of the population a known, non-zero chance of selection, which is what lets researchers calculate margins of error. Snowball sampling gives up that guarantee in exchange for reaching people a random draw could never find.
Techniques
Snowball sampling comes in three main forms, depending on how many referrals each participant gives:
- Linear: a straight-line chain. One subject gives one referral, that referral gives one more, and the chain continues until the sample is large enough.
- Exponential non-discriminative: each subject gives multiple referrals, and each of those provides more in turn, so the sample grows geometrically until there are enough participants.
- Exponential discriminative: subjects again give multiple referrals, but researchers recruit only one referral from each, chosen to fit the study’s objectives.
How Snowball Sampling Works
- First, researchers will form an initial sample by drafting any potential subjects from a population (seeds).
- Even if only a couple of subjects are found at first, researchers will ask those subjects to recruit other individuals for the study. They recruit subjects by encouraging them to come forward on their own. Study participants will only provide specific names of recruited individuals if there is no risk of embarrassment or a violation of privacy. Otherwise, study participants do not identify any names of other potential participants.
- Current participants will continue to recruit others until the necessary sample size has been reached.
Researchers judge when the sample is large enough by saturation, not by a target number picked in advance. Saturation is the point at which new referrals stop producing fresh themes or contacts.
Guest, Bunce and Johnson (2006) found the basic pattern of themes was often in place after around twelve interviews.
Ethical Considerations
Snowball sampling requires special approval by an Institutional Review Board (IRB), whereby the researchers must provide a valid justification for using this method.
Researchers must also take precautions to protect the privacy of potential subjects, especially if the topic is sensitive or personal, such as studies of networks of drug users or prostitutes.
In addition, each respondent has the opportunity to participate or decline. Current participants in this method receive no compensation for referrals. They are also not required to name other potential participants.
Example Situations
Snowball sampling helps when participants are hard to find. It works well for hidden populations, such as criminals, drug dealers, or sex workers, who are difficult for researchers to access directly.
Consider a researcher studying undocumented immigrants in one city. Fear of legal repercussions and a lack of formal records make this population especially hard to reach through traditional sampling.
The researcher starts small. They contact a local organization that supports immigrants and connect with a few willing individuals through it.
These initial participants, the seeds, are then asked to refer the researcher to other undocumented immigrants they know. Each new participant refers others in turn. The sample grows through this chain of referrals, much like a snowball rolling downhill.
The method works well because participants tend to know others like themselves. Members of hidden populations are often closely connected. They share interests or belong to the same groups, so they can vouch for the study and reassure others about confidentiality.
Research Examples
- Researching non‐heterosexual women using social networks (Browne, 2005).
- Investigating lifestyles of heroin users (Kaplan, Korf, & Sterk, 1987).
- Using Facebook to virtually chain-refer Argentinian immigrant entrepreneurs in Spain, an early example of social-media-based snowball sampling (Baltar & Brunet, 2012).
- Studying illegal drug users over the age of 40 (Waters, 2015).
- Obtaining samples of populations at risk for HIV (Kendall et al., 2008).
Advantages
- Reaches hidden populations: concealed groups such as drug users or sex workers are hard to contact directly. A trusted referral from someone already in the study helps new contacts feel safe joining.
- Lower ethical risk: because each respondent can choose to participate or decline, and the method needs IRB approval (see Ethical Considerations), it carries built-in consent and oversight.
- Saves money and time: since current participants locate the next ones, researchers spend less time and money than methods that need a complete list of the population up front.
Limitations
Difficult to determine sampling error
Snowball sampling is a non-probability sampling method, so researchers cannot calculate the sampling error. Each participant’s chance of being included depends on who referred them, not on a known probability.
This breaks the assumption of independence between cases. Standard formulas for sampling error assume independence. Without it, there is no valid way to estimate how far the sample might depart from the true population value.
Bias is possible
Since current participants select other members for the sample, bias is likely. People tend to recommend others like themselves, from within their own social circle.
This means a snowball sample often ends up more homogeneous than the wider population it is meant to represent. The initial participants shape the whole chain that follows. A network of sociable, well-connected people will snowball very differently from a network of more isolated individuals.
Because researchers are not selecting participants themselves, they also have little control over the sample. They typically have minimal knowledge of whether the achieved sample is representative of the target population.
Key Terms
- A sample is the participants you select from a target population (the group you are interested in) to make generalizations about. As an entire population tends to be too large to work with, a smaller group of participants must act as a representative sample.
- Representative means the extent to which a sample mirrors a researcher’s target population and reflects its characteristics (e.g. gender, ethnicity, socioeconomic level). In an attempt to select a representative sample and avoid sampling bias (the over-representation of one category of participant in the sample), psychologists utilize various sampling methods.
- Generalisability means the extent to which their findings can be applied to the larger population of which their sample was a part.
Key Takeaways
- Chain Referral: Snowball sampling starts with a small group of participants who recruit others they know, so the sample grows through personal referral chains.
- Hidden Populations: It is especially useful for reaching populations with no sampling frame, such as drug users, undocumented migrants, and other hard-to-reach or stigmatised groups.
- Non-Probability: Because participants are not randomly selected, snowball sampling cannot support the same statistical inference as random sampling methods.
- Homogeneity Bias: Referred participants often resemble those who recruited them, since people tend to recommend others from their own social circle.
- No Sampling Error: The referral chain breaks the independence assumption behind standard statistical formulas, so researchers cannot calculate a meaningful sampling error for a snowball sample.
- Qualitative Fit: Because representativeness is not the goal, qualitative researchers judge a snowball sample by saturation: whether new referrals keep surfacing fresh information.
- Origins: The method was given its formal statistical treatment by Goodman (1961) and developed further by Biernacki and Waldorf (1981) as a tool for studying concealed populations.
Sources
Baltar, F., & Brunet, I. (2012). Social research 2.0: Virtual snowball sampling method using Facebook. Internet Research, 22 (1), 57–74. https://doi.org/10.1108/10662241211199960
Biernacki, P., & Waldorf, D. (1981). Snowball sampling: Problems and techniques of chain referral sampling. Sociological Methods & Research, 10 (2), 141–163. https://doi.org/10.1177/004912418101000205
Browne, K. (2005). Snowball sampling: Using social networks to research non-heterosexual women. International Journal of Social Research Methodology, 8 (1), 47–60. https://doi.org/10.1080/1364557032000081663
Felix-Medina, M. H., & Thompson, S. K. (2004). Combining link-tracing sampling and cluster sampling to estimate the size of hidden populations. Journal of Official Statistics, 20 (1), 19–38.
Goodman, L. A. (1961). Snowball sampling. The Annals of Mathematical Statistics, 32 (1), 148–170. https://doi.org/10.1214/aoms/1177705148
