Snowball
Sampling
Abstract
Snowball sampling is a non-probability
sampling technique commonly used in social sciences to identify research
participants from hard-to-reach or hidden populations. The method begins with a
small group of initial respondents who then refer additional participants from
their network. This creates a “snowball” effect, helping the researcher gather
a larger sample through referrals. The technique is useful when sampling frames
are unavailable, though it may lead to bias due to the non-random nature of
selection.
Keywords
Snowball sampling, referral sampling,
non-probability sampling, hidden
populations, chain sampling.
Introduction
Sampling is a crucial aspect of research,
especially when the target population is not easily identifiable or accessible.
Traditional probability sampling may not be suitable for studying rare,
sensitive, or hidden populations. In such cases, snowball sampling serves as an
effective strategy by using social networks to reach participants. This method
starts with a few initial respondents who refer others, enabling the sample to
expand gradually.
The given Figure 1 provides a visual representation of snowball sampling
Figure 1
Snowball Sampling
From
Figure 1 Snowball Sampling can be easily comprehended
Meaning of Snowball Sampling
Snowball sampling is a
chain-referral sampling method in which existing participants recruit new
participants from their acquaintances. The sample grows larger as more
individuals are added through referrals, resembling a snowball increasing in
size.
Characteristics of Snowball Sampling
Snowball sampling is
particularly useful when the target population is hidden, rare, or difficult to
identify, making conventional sampling methods less effective (Bryman, 2016).
In this method, recruitment occurs through referrals made by existing participants,
enabling researchers to reach individuals who may otherwise remain inaccessible
(Kothari, 2004). The sampling process continues until the researcher obtains
the required sample size, with participants progressively nominating others in
their network (Goodman, 1961). Snowball sampling is frequently used in
qualitative studies because it allows deeper access to specific groups,
although it may also be applied in quantitative research involving
hard-to-reach populations (Creswell, 2014). However, because the technique is
non-random and dependent on social networks, it may result in limited
generalizability and potential sampling bias (Bryman, 2016).
Steps in Snowball Sampling
Step 1:
Identify the Target Population
The researcher first defines and understands the group that
needs to be studied, especially if they are hidden or hard to reach.
Step 2: select
initial participants (seeds)
A small number of individuals belonging to the target
population are selected to begin the sampling process.
Step 3: Collect
Data From Initial Participants
Interviews, surveys, or observations are conducted with the
initial subjects.
Step 4: Ask For Referrals
Participants Are Requested to Share Information About Others
Who Match The Study Criteria.
Step 5: Recruit
New Participant. The referred individuals are contacted and included in the study
if they meet the criteria and consent to participate.
Step 6:
Continue the Referral Chain
Each new participant refers additional individuals, causing
the sample to expand.
Step 7: Reach
Saturation or Desired Sample Size
Types of snowball
sampling are given below in Table 1
Table 1
Types of Snowball Sampling
Type of Snowball Sampling | Description
| Growth Pattern | When to Use |
Linear snowball sampling | Each participant refers only one new participant. | Slow, single- chain expansion. | When the researcher wants controlled, step by-step recruitment.
|
Exponential non- discriminative snowball sampling | Each participant refers multiple new participants, and all referred individuals are included.
| Rapid expansion— sample grows quickly like a branching tree. | When reaching a large hidden population quickly is needed. |
Exponential discriminative snowball sampling | Participants refer multiple individuals, but the researcher selects only some based on criteria. | Moderate, selective expansion | When the researcher needs faster recruitment but with controlled participant selection. |
From Table 1 it is evident that each type of snowball
sampling varies in referral structure, growth pattern, and suitability for
different research purposes.
Advantages of Snowball Sampling
Snowball sampling offers several advantages, especially in
studies involving hidden, rare, or vulnerable populations. It is highly
effective for accessing groups that are difficult to reach through conventional
sampling methods because referrals from known individuals build credibility and
trust (Bryman, 2016). The method also saves time and resources, as participant
recruitment progresses naturally through social networks rather than extensive
field searching (Creswell, 2014). Peer referrals help establish rapport more
efficiently, allowing researchers to collect deeper and more reliable data
(Goodman, 1961). Additionally, snowball sampling is particularly suitable for
exploratory and qualitative research, where flexibility and access to insider
networks are essential (Kothari, 2004).
Limitations of Snowball Sampling
Despite its
usefulness, snowball sampling includes several limitations that researchers
must consider. One significant issue is the high risk of sampling bias, as
participants are selected through social networks rather than random methods
(Bryman, 2016). This often results in the overrepresentation of individuals who
are socially connected, while isolated members of the population may remain
excluded (Goodman, 1961). Consequently, the findings have limited
generalizability, particularly in quantitative studies where representativeness
is crucial (Creswell, 2014). Ethical concerns may also arise, especially
regarding privacy and confidentiality, since participants disclose information
about others during the referral process (Kothari, 2004).
Conclusion
Snowball sampling is a
valuable technique for studying populations that are not easily accessible
through traditional sampling methods. By relying on referrals from initial
participants, this method helps researchers reach wider networks. Despite
limitations such as bias and restricted generalizability, snowball sampling
remains highly effective in qualitative and exploratory research where trust
and access are essential.
References
Bryman, A. (2016). Social research methods (5th ed.).
Oxford University Press.
Creswell, J. W. (2014). Research design: Qualitative,
quantitative, and mixed methods approaches (4th ed.). SAGE Publications.
Goodman, L. A. (1961). Snowball sampling. Annals of
Mathematical Statistics, 32(1), 148–170. https://doi.org/10.1214/aoms/1177705148
Kothari, C. R. (2004). Research methodology: Methods and
techniques (2nd ed.). New Age International Publishers.
