Snowball Sampling

A non-probability sampling technique in which researcher begins with a small population of known individuals and expands the sample by asking those initial participants to identify others that should participate in the study. In other words, the sample starts small (like a snowball running down the hill) but into a larger sample through the course of the research.Snowball sampling is also known as chain sampling, cold-calling, chain-referral sampling or referral sampling.

This often occurs when the population is somehow marginalized, like homeless or formerly incarcerated individuals or those who are involved in illegal activities. It is also common to use this sampling technique with people whose membership in a particular group is difficult to find not widely known, such as closeted gay people or bisexual or transgender individuals, homeless people, drug addicts, members of an elite golf club etc.Some people may not want to be found.

Example:

If a study requires investigating cheating on exams, shoplifting, drug use, or any other “unacceptable” social behavior, the participants would worry to come forward due to possible ramifications.

However, other study participants would likely know other people in the same situation as themselves and could inform others about the benefits of the study and reassure them of confidentiality.

Types of Snowball Sampling:

Linear snowball sampling:

Forming a sample group which starts with only one subject and the subject refers only one referral and further continues until the fully sampled group is formed.

Exponential non-discriminating snowball sampling:

The first subject recruited to the sample group refers one or more referrals. Each new referral is explored until primary data forms sufficient amount of samples.

Exponential discriminating snowball sampling:

Subjects give multiple referrals but only one subject is recruited among them to form effective samples.

Snowball sampling consists of two steps:

Identify potential subjects from sampling frame in the population.Mostly one or two subjects can be found at this stage.

Ask those subjects to recruit other people and then ask those people to recruit till there are no more subjects left/sample size becomes unmanageable.

Advantages:

This process allows the researcher to reach populations that are difficult to sample compared to other sampling methods.

The process is cheap, simple and cost-efficient.

This technique needs lesser workforce as the subjects were involved directly compared to other sampling techniques.

 Disadvantages:

Oversampling a particular network of peers may lead to biasing.

There is no guarantee about the representation of samples. It is not possible to determine the actual pattern of distribution of population.

Determination of the sampling error and make statistical inferences from the sample to the population is not possible due to the absence of random selection of samples.